I'm not saying "humongous" slowdowns. But if it is interpreted/compiled at runtime, that would be bad. That was the big problem Apple had with the 68xx to PPC conversion, interpreted code. I think it was Connectix that stored the code and fixed most of the performance problems You might not remember, or weren't old enough at the time, but 68xx programs ran at about half the speed they ran on the 68xx.
Well, I guess you're definition of bloat and my definition of bloat are two separate things. To me, a bloated application is one that has so many feauters, options, abilities, etc, that it struggles under its own weight. For example, people whine about Mozilla as a bloated application, because it contains a mail client, composer, news reader, etc, as well as a browser. A large application, by default, is NOT bloated. Its just big. And big isn't a big deal with the size of harddisks these days. With separate binaries for x86 and PPC, only half the code is read, anyway, so the other stuff just sits there unused AND unloaded. I never heard anyone complain about Fat apps being bloated when the PPC came out, because people understood that some code would be ignored, and other run, depending on the machine. (Now people complained the files were big, and went through some effort to delete the PPC part if they didn't need it, because disk space was much more of a premium, and much more expensive per megabyte, then).
Also, nothing says an app has to be both a PPC and Intel app. Some 'fat' apps (that ran both on the PPC and 680x0) came in pairs, one for PPC and one for 680x0. But the only advantage here was disk space. Adobe's installer, for example, could be set up to install either binary, rather than both. However, the theory is that they'd need just one set of code to be compiled for each platform (note: no comment here on length of making it compatible for either).
As for the 680x0 emulator in System 7 and beyond, yes, I can somewhat remember that (though I'm old enough to have used my father's 128K Mac back in 1984 for college, so take that you yung'un!). And the emulation wasn't THAT bad. Its performance was offset by the fact the PPC was much faster than the 680x0. And Apple ended up rewriting the emulator to something like connectix's (storing rather than just interpreting and tossing). Some code was incredibly slow (esp floating point, which required Software FPU to emulate, hey, I remember that too!).
[Side point: 10+ years later, in Classic and all, the emulator still exists, and there's large hunks of System 9 that is still written in 680x0 code and still being emulated.]
But I'm pretty sure we're not talking run-time interpreting/compiling here. [Another note: .Net's CLR compiles at run-time, although it actually compiles and stores the code up-front, so loading is slower, but running is just as quick as native, well, not as quick as native, as quick as any MS software can run, I guess. But they seem to get good performance on the PC. Its the Mac where it lags]
If that's the case then Longhorn will be out in no time.
SJ is going headlong against Longhorn.
SOO TRUE WINDOWsusks, but I think they have been testing it for quite a while on Intel and MANY people in the industry now about this, and probably dread it, wish it weren't true.
They dread it because they realize that winWows has never had any real competition, not like the competition Apple will create !!!
I have been a Mac and Windows test lead at almost every place I have worked at and when I am job hunting my result are very strange, and many bazar things have happened to me in this area.
My recomdation is if you have Mac and Intel experience make two resumes, one for each do not both with a mixed resume.
I think it would be VERY WISE to make the new Mac OS Xntel MACchines winWows compatibo !!!
I think it would be VERY WISE to make the new Mac OS Xntel MACchines winWows compatibo !!!
Have you been drinking again?
I cannot see Apple doing x86 Intel chips...unless they are going to announce a product(s) tomorrow and also announce a date within the reasonable future that they are shipping. That's the thing that just plain doesn't sit right with me. If that were settled like so, I might be more persuaded...
I cannot see Apple doing x86 Intel chips...unless they are going to announce a product(s) tomorrow and also announce a date within the reasonable future that they are shipping. That's the thing that just plain doesn't sit right with me. If that were settled like so, I might be more persuaded...
Being winWows compatibo would be the best thing yet.
Then Apple does NO WORK to support winWows other then perhaps building the hardware to duel boot.
And peopo run the two side by side and learn all the wows of winWows.
They could announce this tommorrow as it would effect software a great deal, but normally I would say you would not hear this until it ships.
If its true I bet they give demos with a duel core Intel laptop MACchine.
Well you can imagine all you like, but in reality the number of issues that arise in a port like this is surprising, even to those of us who have been doing it for decades. If the code was written simultaneously for multiple platforms, it will be easily portable between those platforms. The more platforms, the easier it is to port. Any time a new platform comes along (and OSX on x86 is a new platform), however, you're going to have some troubles. If you don't believe that you simply haven't been programming or porting code long enough.
I was a bit polite. Let me re-iterate. I know that Big endianess is not a problem. People at application level are using apis which they expect will deal with this, and so they will. This includes Caron apis for reading from disk and Cocoa apis for reading from disk ( or even IPC). this is not an issue.
The rest of your post is just junk. OSX is not a new platform on intel. The lower level abstracts for all application developers.
Feel free to write code for the mac using existing i/o API that has big endian problems.
As for credentialism, I have been coding for the Mac , and porting, for a decade. At the very lowest level. I have also worked on pre-release versions of OS X 1.0. Since then.
If this is announced tomorrow then all WWDC developers will have their applications ported with the week ( since it would take a recompile) and most by Tuesday. It is that simple.
I was a bit polite. Let me re-iterate. I know that Big endianess is not a problem. People at application level are using apis which they expect will deal with this, and so they will. This includes Caron apis for reading from disk and Cocoa apis for reading from disk ( or even IPC). this is not an issue.
The rest of your post is just junk. OSX is not a new platform on intel. The lower level abstracts for all application developers.
Feel free to write code for the mac using existing i/o API that has big endian problems.
As for credentialism, I have been coding for the Mac , and porting, for a decade. At the very lowest level. I have also worked on pre-release versions of OS X 1.0. Since then.
If this is announced tomorrow then all WWDC developers will have their applications ported with the week ( since it would take a recompile) and most by Tuesday. It is that simple.
So the transition will be instantaneous and will involve little if no bugs at all? If moving from PPC to x86 is going to this painless why didn't Jobs do it years ago? (and don't blame it all on IBM)
I wonder if this gives more weight to the story/news
Not at all. Take a look at the very wording of the article (emphasis mine):
Quote:
Apple CEO Steve Jobs is expected to announce Monday morning at the company's software developers conference in San Francisco that Apple will discontinue using microprocessor chips made by IBM in favor of Intel chips, according to CNET Networks Inc.'s News.com and The Wall Street Journal.
Officials from Apple, Intel Corp. and International Business Machines Corp. could not be reached Sunday to confirm the report.
If these rumors are true, IBM has shot itself in the foot. Sure the game consoles will make a great band-aid, but here they are on the cusp of a huge rise in Apple Macintosh marketshare and they don't want to develop for a "small volume" company. The idiocy of this is that if they would just produce a wide variety of PPc chips, that Apple would become a arge volume company in short order. Especially if one of those chips is a mobile G5. Powerbooks would probably surpass iPods in sales.
Foolish business logic by IBM. This hurts two companies
I guess IBM has no business ethics. They were just ufing the Apple relationship to get free advertising from Apples outstanding products and waiting for multiple customers to come running who they would then cater to while dissing the "small volume" company who got them there.
If the move to Intel enables my next Powerbook to run faster and as reliable as 2.3 GHZ a Powermac then all is well and good.
Otherwise...
If this rumor is true, I will never buy anything IBM related again. Just on principal and ethics.
If that means no Nintendo, microsoft, or sony, then so be it.
Its a cut-throat world out there.
Steve Jobs would not make the move, though, without a solid plan and without knowing for sure that the Intel procs would be superior to the IBM ones.
The most unbelievable part of an all out transition to X86 processors is the time table. This flaw has been pointed out several times. Announcing such a plan will likely kill sales of current Apple products, as even the most loyal Mac users wait for the X86 generation of hardware.
Two alternative theories have been proposed that seem consistent with the news reports. One is that Intel builds a chip with a PPC front end and a means of handling AltiVec code. The second is that Apple will use Transitive's emulation technology to run PPC applications.
I'd like to propose a third theory. (If it has been mentioned, sorry for the repeat.) Possibly Apple plans to begin a dual platform strategy in a year. Rather than abandon the Power PC, Apple will add the X86, and Apple need not build all the hardware. Apple can approve certain models that will run OS X, for which the manufacture pays a fee. You might say this is a tightly controlled clone program.
Apple may build some X86 models too. For example, if neither IBM nor FreeScale can supply good enough PowerBook/iBook PPC chips, Apple may choose to go X86 for a while with these models. This gives Apple the options Steve likes to have, it seems.
The down side of such a plan, if it is a downside, is that two versions of Applications and the OS will need to be provided basically forever. Developers have a year to get ready for the light duty X86 Macs. In two years, even higher end X86 Macs will appear.
Such a plan would allow Apple to continue to making PPC Macs, using the best that IBM and FreeScale have to offer. Apple is simply adding the X86 to their portfolio of supported products.
Comments
Except that WSJ doesn't as a rule report things unless they are very, very sure about their soures.
This is very, very true. The Wall Street Journal is usually very conservative about its claims...this is likely to be true in some form.
That's a very interesting angle as well.
A couple of weeks ago there was an article that said there was a version of Tiger runnnnnnning on Intel in the labs of Intel and other Intel shops.
I have always tended to believe the rumors that said from the beginning of OS X they have maintained a version running on Intel.
Just because Apple is large enough of a company to do that sort of R&D thing.
I referenced the article in my other posts in the other thread.
Originally posted by MACchine
If these recent reports are true then the others probably are also.
A couple of weeks ago there was an article that said there was a version of Tiger runnnnnnning on Intel in the labs of Intel and other Intel shops.
I have always tended to believe the rumors that said from the beginning of OS X they have maintained a version running on Intel.
Just because Apple is large enough of a company to do that sort of R&D thing.
I referenced the article in my other posts in the other thread.
If that's the case then Longhorn will be out in no time.
Originally posted by melgross
You don't call that bloat? I think most would.
Compatibility; read my reply to "asdasd".
I'm not saying "humongous" slowdowns. But if it is interpreted/compiled at runtime, that would be bad. That was the big problem Apple had with the 68xx to PPC conversion, interpreted code. I think it was Connectix that stored the code and fixed most of the performance problems You might not remember, or weren't old enough at the time, but 68xx programs ran at about half the speed they ran on the 68xx.
Well, I guess you're definition of bloat and my definition of bloat are two separate things. To me, a bloated application is one that has so many feauters, options, abilities, etc, that it struggles under its own weight. For example, people whine about Mozilla as a bloated application, because it contains a mail client, composer, news reader, etc, as well as a browser. A large application, by default, is NOT bloated. Its just big. And big isn't a big deal with the size of harddisks these days. With separate binaries for x86 and PPC, only half the code is read, anyway, so the other stuff just sits there unused AND unloaded. I never heard anyone complain about Fat apps being bloated when the PPC came out, because people understood that some code would be ignored, and other run, depending on the machine. (Now people complained the files were big, and went through some effort to delete the PPC part if they didn't need it, because disk space was much more of a premium, and much more expensive per megabyte, then).
Also, nothing says an app has to be both a PPC and Intel app. Some 'fat' apps (that ran both on the PPC and 680x0) came in pairs, one for PPC and one for 680x0. But the only advantage here was disk space. Adobe's installer, for example, could be set up to install either binary, rather than both. However, the theory is that they'd need just one set of code to be compiled for each platform (note: no comment here on length of making it compatible for either).
As for the 680x0 emulator in System 7 and beyond, yes, I can somewhat remember that (though I'm old enough to have used my father's 128K Mac back in 1984 for college, so take that you yung'un!). And the emulation wasn't THAT bad. Its performance was offset by the fact the PPC was much faster than the 680x0. And Apple ended up rewriting the emulator to something like connectix's (storing rather than just interpreting and tossing). Some code was incredibly slow (esp floating point, which required Software FPU to emulate, hey, I remember that too!).
[Side point: 10+ years later, in Classic and all, the emulator still exists, and there's large hunks of System 9 that is still written in 680x0 code and still being emulated.]
But I'm pretty sure we're not talking run-time interpreting/compiling here. [Another note: .Net's CLR compiles at run-time, although it actually compiles and stores the code up-front, so loading is slower, but running is just as quick as native, well, not as quick as native, as quick as any MS software can run, I guess. But they seem to get good performance on the PC. Its the Mac where it lags]
Originally posted by brent1a
If that's the case then Longhorn will be out in no time.
I was thinking, well then, where are the weapons of mass destruction?
Originally posted by DHagan4755
I was thinking, well then, where are the weapons of mass destruction?
It took 7 pages to get me to laugh but it finally happened.......
Originally posted by brent1a
If that's the case then Longhorn will be out in no time.
SJ is going headlong against Longhorn.
SOO TRUE WINDOWsusks, but I think they have been testing it for quite a while on Intel and MANY people in the industry now about this, and probably dread it, wish it weren't true.
They dread it because they realize that winWows has never had any real competition, not like the competition Apple will create !!!
I have been a Mac and Windows test lead at almost every place I have worked at and when I am job hunting my result are very strange, and many bazar things have happened to me in this area.
My recomdation is if you have Mac and Intel experience make two resumes, one for each do not both with a mixed resume.
I think it would be VERY WISE to make the new Mac OS Xntel MACchines winWows compatibo !!!
Originally posted by MACchine
I think it would be VERY WISE to make the new Mac OS Xntel MACchines winWows compatibo !!!
Have you been drinking again?
I cannot see Apple doing x86 Intel chips...unless they are going to announce a product(s) tomorrow and also announce a date within the reasonable future that they are shipping. That's the thing that just plain doesn't sit right with me. If that were settled like so, I might be more persuaded...
Originally posted by DHagan4755
I was thinking, well then, where are the weapons of mass destruction?
The provocation for the search for weapons of Mass Dilution was the arms inspections.
The whole world was in stitches with Soddama over his manipulations of the arms inspectors.
IF THERE WERE NO WEAPONS OF MASS CONSTIPATION THEN WHY DID HE MOLEST THE INSPECTORS ???
Kerry said he would have gone to war and he voted for war, thus your rehistorations are true INSANITY !!!
Originally posted by DHagan4755
Have you been drinking again?
I cannot see Apple doing x86 Intel chips...unless they are going to announce a product(s) tomorrow and also announce a date within the reasonable future that they are shipping. That's the thing that just plain doesn't sit right with me. If that were settled like so, I might be more persuaded...
Being winWows compatibo would be the best thing yet.
Then Apple does NO WORK to support winWows other then perhaps building the hardware to duel boot.
And peopo run the two side by side and learn all the wows of winWows.
They could announce this tommorrow as it would effect software a great deal, but normally I would say you would not hear this until it ships.
If its true I bet they give demos with a duel core Intel laptop MACchine.
IF THERE WERE NO WEAPONS OF MASS CONSTIPATION THEN WHY DID HE MOLEST THE INSPECTORS ???
I would guess for the same reason Bill Gates has molested the world for so long............what? Why did I provoke this?
Originally posted by brent1a
I would guess for the same reason Bill Gates has molested the world for so long............what? Why did I provoke this?
Umm, Intel used to act a lot like Soddam0m.
And MS like weapons inspectors.
Maybe Intel's new CEO is more like G Bush !?!?!?!?!
Well you can imagine all you like, but in reality the number of issues that arise in a port like this is surprising, even to those of us who have been doing it for decades. If the code was written simultaneously for multiple platforms, it will be easily portable between those platforms. The more platforms, the easier it is to port. Any time a new platform comes along (and OSX on x86 is a new platform), however, you're going to have some troubles. If you don't believe that you simply haven't been programming or porting code long enough.
I was a bit polite. Let me re-iterate. I know that Big endianess is not a problem. People at application level are using apis which they expect will deal with this, and so they will. This includes Caron apis for reading from disk and Cocoa apis for reading from disk ( or even IPC). this is not an issue.
The rest of your post is just junk. OSX is not a new platform on intel. The lower level abstracts for all application developers.
Feel free to write code for the mac using existing i/o API that has big endian problems.
As for credentialism, I have been coding for the Mac , and porting, for a decade. At the very lowest level. I have also worked on pre-release versions of OS X 1.0. Since then.
If this is announced tomorrow then all WWDC developers will have their applications ported with the week ( since it would take a recompile) and most by Tuesday. It is that simple.
Originally posted by asdasd
I was a bit polite. Let me re-iterate. I know that Big endianess is not a problem. People at application level are using apis which they expect will deal with this, and so they will. This includes Caron apis for reading from disk and Cocoa apis for reading from disk ( or even IPC). this is not an issue.
The rest of your post is just junk. OSX is not a new platform on intel. The lower level abstracts for all application developers.
Feel free to write code for the mac using existing i/o API that has big endian problems.
As for credentialism, I have been coding for the Mac , and porting, for a decade. At the very lowest level. I have also worked on pre-release versions of OS X 1.0. Since then.
If this is announced tomorrow then all WWDC developers will have their applications ported with the week ( since it would take a recompile) and most by Tuesday. It is that simple.
So the transition will be instantaneous and will involve little if no bugs at all? If moving from PPC to x86 is going to this painless why didn't Jobs do it years ago? (and don't blame it all on IBM)
Originally posted by amac4me
I wonder if this gives more weight to the story/news
Not at all. Take a look at the very wording of the article (emphasis mine):
Apple CEO Steve Jobs is expected to announce Monday morning at the company's software developers conference in San Francisco that Apple will discontinue using microprocessor chips made by IBM in favor of Intel chips, according to CNET Networks Inc.'s News.com and The Wall Street Journal.
Officials from Apple, Intel Corp. and International Business Machines Corp. could not be reached Sunday to confirm the report.
Repeat after me: ECHO CHAMBER
Foolish business logic by IBM. This hurts two companies
I guess IBM has no business ethics. They were just ufing the Apple relationship to get free advertising from Apples outstanding products and waiting for multiple customers to come running who they would then cater to while dissing the "small volume" company who got them there.
If the move to Intel enables my next Powerbook to run faster and as reliable as 2.3 GHZ a Powermac then all is well and good.
Otherwise...
If this rumor is true, I will never buy anything IBM related again. Just on principal and ethics.
If that means no Nintendo, microsoft, or sony, then so be it.
Its a cut-throat world out there.
Steve Jobs would not make the move, though, without a solid plan and without knowing for sure that the Intel procs would be superior to the IBM ones.
_peace
The most unbelievable part of an all out transition to X86 processors is the time table. This flaw has been pointed out several times. Announcing such a plan will likely kill sales of current Apple products, as even the most loyal Mac users wait for the X86 generation of hardware.
Two alternative theories have been proposed that seem consistent with the news reports. One is that Intel builds a chip with a PPC front end and a means of handling AltiVec code. The second is that Apple will use Transitive's emulation technology to run PPC applications.
I'd like to propose a third theory. (If it has been mentioned, sorry for the repeat.) Possibly Apple plans to begin a dual platform strategy in a year. Rather than abandon the Power PC, Apple will add the X86, and Apple need not build all the hardware. Apple can approve certain models that will run OS X, for which the manufacture pays a fee. You might say this is a tightly controlled clone program.
Apple may build some X86 models too. For example, if neither IBM nor FreeScale can supply good enough PowerBook/iBook PPC chips, Apple may choose to go X86 for a while with these models. This gives Apple the options Steve likes to have, it seems.
The down side of such a plan, if it is a downside, is that two versions of Applications and the OS will need to be provided basically forever. Developers have a year to get ready for the light duty X86 Macs. In two years, even higher end X86 Macs will appear.
Such a plan would allow Apple to continue to making PPC Macs, using the best that IBM and FreeScale have to offer. Apple is simply adding the X86 to their portfolio of supported products.