Apple to Use Intel Chips?

Stack_13 writes "Wall Street Journal reports that Apple will agree to use Intel chips. Neither Apple or Intel confirm this. Interestingly, PCMag's John C. Dvorak predicted this for 2004-2005. Are even cheaper Mac Minis coming?"

Comments

Does this mean -

[thewldisntenuff]

We'll see Mac OS X - x86 anytime soon?

Re:Does this mean -

[southpolesammy]

More likely it will mean that you'll see better pricing on PowerPC-based Macs in the future.

Re:Does this mean -

[rovingeyes]

Good point. But is Apple a software company or a hardware company? If OS X-86 come out then they have to dedicate an entire department for that. Not that Apple cannot afford it, but if they are really a hardware company, it really is not lucrative enough and not to mention inheriting all the troubles of x86 world.

Re:Does this mean -

[taskforce]

It was rumoured a while back that Apple had an x86 internal build of OS X. It's not unlikely becuase of it's FreeBSD roots that it was part of the development cycle at least for the early versions of OS X. (I don't see why they would continue to develop the later versions on x86 if they wheren't planning this all along)

Re:Does this mean -

[selderrr]

I seriously doubt it. Right now, Intel is not really leading he pack in processor terms. All you hear about these days is IBM (both PS3 and XBox360 are IBM powered) and AMD on the 64bit front. With OSX being the most widely spread 64bit consumer OS, I can hardly believe Apple switching to the least-represented CPU manufacturer.

If anything, they could perhaps use some non-x86 intel stuff for portables devices where the PPC sucks up too much power. Remember that Intel is more than only x86

Re:Does this mean -

[Oculus Habent]

This could be the same tactic Dell uses with Intel... "We could go with AMD, but about those prices..."

Cheaper because of Intel? I doubt it. Even if Apple does start using x86 - or more likely x86-64 - they would still likely use their own controller chips (Note that Apple uses a single, integrated controller rather than a north/southbridge approach) and custom boards.

It's not impossible that Apple will switch to Intel processors. We already know they keep a copy of the OS up to date on Intel hardware, and even released Darwin x86. The problems come from all the things they would leave behind:

Compatibility - The PowerPC architecture emulates x86 better than the other way 'round. To keep from eliminating all old software with one fell swoop, they would need to emulate PowerPC. This would cause old software to run like death.

VMX - Much of Apple's current power comes from the AltiVec/VMX/Velocity Engine available on the G4 & G5 processors. It is what offers Apple serious performance benefits in certain applications, and makes possible many of the near/realtime capbilities in programs like iPhoto, iMovie, and even Final Cut Pro. Unless Intel tacks on a VMX unit, I don't see Apple switching.

Maybe a dual-processor system: one PowerPC and one Intel? Not likely, I grant you.

Re:Does this mean -

[Anonymous Coward]

The thing that sets Apple apart from all other companies in this area is that they aren't just a hardware company or a software company. They are both. Most people buy the hardware because of the excellent software they offer on top. It's the combined experience that makes their hardware stand above the rest.

Re:Does this mean -

[/ASCII]

Apple may be planning on using Intel network cards. Or maybe one of intels hardware raid chips. Flash memories, Cellular processors, wireless chips are al possible. But processors? I doubt it.

Re:Does this mean -

[defy god]

safari, iWork, final cut, iCal, iChat, iTunes, iMovie, iDVD (not to mention the PRO versions), MS office (yeah, right..), Photoshop, Illustrator.. okay.. you get the point.

Re:Does this mean -

[/ASCII]

My guess is they really are planning on using Intel chips - just not processors. Remember, Intel produces wireless chips, Flash memory, Ethernet chips, and Salt and Vinegar chips.

Re:Does this mean -

[garcia]

This has been discussed before. Apple uses their x86 kernel as leverage against MSFT so that there is still Office for OS X.

The second that Apple moves into the market with OS X for x86, MSFT is going to pull Office and render OS X basically useless compared to Windows. Yeah, there are open alternatives that sorta work but in the real world people want to use what they are comfortable with. Unfortunately that's Office.

MSFT knows that if they pulled Office for OS X that Apple could easily release OS X for x86 and enter a new competitor into the OS market.

Re:Does this mean -

[squiggleslash]

You'd be surprised how architecturally different Apple's regular offerings are from one another. I believe even Apple's latest PowerBooks use technologies considered obsolete in the rest of the line, such as ADB for the internal bus used for the keyboard and pointing device. The G4 and G5s have much bigger differences between them than the G3s and G4s, and Apple is trying to support a whole range of systems from the ground up.

In that respect, it may be easier for Apple to switch to an entirely new CPU architecture than you might think. The additional support wouldn't be dramatic, it could continue to have a lot in common with the rest of their systems (which heavily use USB and IDE, PCI and AGP, etc), making the CPU and a few other minor details the major changes. It certainly wouldn't need a dedicated department of any serious size to support this version of OS X, it'd just be an additional platform to test the recompiled version upon.

This is, of course, assuming we're talking about Intel chips being used in Macs (with an OS X compiled to run on it) and not a generic version of OS X being developed that'll run on IBM PC clones, which is an entirely different issue.

Re:Does this mean -

[geoffspear]

Please explain the process whereby Apple will convert everyone's old applications into fat binaries. Without access to the source.

Re:Does this mean -

[Oculus Habent]

Fat binaries are great for the transition phase, but don't do anything for old apps. If I just started my DTP company and plunked down $7500 for various software packages, I would not be happy to hear that none of it will run on the next Mac I buy. Just as they emulated the 680x0 on the PowerPC - which is still available under Classic - they would need to emulate the PowerPC under the x86.

Re:Does this mean -

[Klivian]

>Cheaper because of Intel? I doubt it.
Exactly, the big price difference between Apple and generic x86 hardware are not mainly caused by the price of the processor. The price difference lays in the commodity hardware contra Apples custom boards, and the resulting competition where price are one of the driving factors. Since Apple does not have competition on their hardware they can chose a much more comfortable price margin on their hardware. Given that Apple are a hardware company any possible shift to x86 will not give significant lower prices, as Apple still will not open the platform to clone makers.

Besides Apple has already changed processor architecture once, so they have experience with the process involved.

Re:Does this mean -

[Halo1]

The gigabit ethernet chip in my old G4/400 in fact is an Intel chip.

Re:Does this mean -

[/ASCII]

Yeah that really sets Apple apart from other companies like Sun, IBM and HP... NOT. All the Unix providers have exactly the same control.

You got it backwards. There are many closed architectures with one company dictating hardware and software. It is in fact the x86 that is unique in that multiple companies provide each part of the computer in an open architecture. And though this solution has it's problems, I think it has shown itself to be vastly superior to a closed system like the Mac.

Also keep in mind that if Apple where the dominating computer provider, they could probably had squished open source efforts like Linux in the cradle by closing specs and making new hardware incompatbile. The X86 may not be pretty, but I'd prefer it over a closed architecture any day.

Re:Does this mean -

[shaitand]

The transition to powerpc was not a change of architecture, it was an upgrade in architecture. x86 is an entirely different architecture with an entirely different instruction set.

Not to mention, the PowerPC processor is the only edge Macs have left on PC hardware. If Apple goes x86 the Mac will simply be an overpriced PC running a pretty gui on top of BSD.

Re:Does this mean -

[Shisha]

Valid point. They won't of course. I've been assuming that the major players would release appropriate versions sometime in the future. I know there are problems with this:
- Quark Xpress might take ages (again, remember how long OS X version took).
- Microsoft might actually not want to release office at all.

(actually maybe they could convert things on the fly and then cache the results, sort of a better emulation process... there is some mention of something along those lines on macrumors.com. It'll be a bit like running Java bytecode. They could then profile the things on the fly and optimize the most used parts of the program... JVM do all sorts of clever trickery nowdays.)

Anyway this is all purely hypothetical. Apple's not moving to x86 CPUs. As other people said, it's probably some random chip they might want from Intel.

Re:Does this mean -

[Kyro]

The latest powerbooks actually have a USB interface for keyboard and mouse. however you are correct, the ibooks, and all previous powerbooks used ADB.

Re:Does this mean -

[cirisme]

>> If you could buy the parts and build your own Mac they'd be alot more appealing to people...

No, it wouldn't. A lot more appealing to geeks, perhaps, but not people in general. Most people don't build their own systems, but buy the cheapest crap from Dell, HP, Gateway, etc.

Re:Does this mean -

[chasingporsches]

nope. you might see intel making powerpc though. if apple ports mac os x to x86, thats cool and all, but none of the applications will work, then they'll go back to the mess of the 68k to powerpc switchover, where you have 2 versions of every product. considering powerpc is working so well for apple, and steve jobs just said he's happy with the powerpc architecture, then this story seems to be bogus. not worth apple's effort.

Re:Does this mean -

[Anonymous Coward]

Going out on a limb here: yes.

Keeping my feet on the ground here: No.

We will see Windows on PowerPC long before we ever see the full OS X on x86. There's absolutely no advantage to changing platforms at this point.

Sure, the Intel/AMD world looked very attractive when Apple was relying on Motorola and lagging way behind in CPU speeds, but current PowerPC technologies from IBM are outstanding. The G5 is a terrific chip. Multi-core PowerPC chips offer a great deal of promise in the very near future.

If Apple does move away from the G5 archetecture, it will be to go to Cell chips, not Intel-64.

Re:Does this mean -

[lokedhs]

If you could buy the parts and build your own Mac they'd be alot more appealing to people...

In so many words: No, they wouldn't.

Very few people build their own computers. Most buy a box from their local computer store, or order it online from a company like Dell.

Also, no one would run OSX on a standard PC. Just like no one runs BeOS, or ran OS/2. An x86 Apple would probably be a proprietary Apple with an x86 in, and no one would care. Just like no one cares that there is a PowerPC in them today.

Re:Does this mean -

[chasingporsches]

or Intel XScale processors for a video ipod...

Re:Does this mean -

[NMerriam]

Whether you call it an "upgrade" or a change is semantics. The PPC and 680x0 had different instruction sets and required completely different programming at the system level -- that Apple built 680x0 system-level software emulation (and later on-the-fly dynamic recompilation) and made it completely transparent to the end-user was a pretty significant feat.

Not to mention, the PowerPC processor is the only edge Macs have left on PC hardware. If Apple goes x86 the Mac will simply be an overpriced PC running a pretty gui on top of BSD.

Whatever. When Ferarri bulds a car with an automatic transmission, it's just an overpriced Taurus with a pretty body kit, right?

After all, what kind of crazy computer USER would buy a computer based on the USER interface? Everybody knows your decision should be based on whether the system is little-endian or big-endian!

Re:Does this mean -

[quelrods]

By closed architectures are you referring to the PPC? This is not produced by Apple but by IBM and the architecture docs are quite good (IBM will mail you a hardcopy set of the books for free.) In fact AMD and Intel for the x86 have docs on par with IBM for the PPC. I think you meant to say that some of their peripherials are closed products (like their wireless.)

Re:Does this mean -

[arloguthrie]

Most notably, the XServe Raid runs on an Intel processor. There are tons of reasons Apple would be meeting with Intel. One day, perhaps "the news" will actually be news and not gossip based on "a friend-of-a-friend told me..."

Re:Does this mean -

[cloudmaster]

Yeah, macs are no good for personal computing. If they used Intel chips, then they'd be personal computers; but since they use PPC chips, macs are only suited for server and mainframe usage.

Re:Does this mean -

[Inkieminstrel]

I think the GP meant that despite the fact that x86 is a bad architecture, it is the dominant one, and it is dominant because the market favors a bad open architecture over a good closed one.

In this case superior means which one will win in the market, not which one is the best design-wise.

Re:Does this mean -

[Tenebrious1]

I seriously doubt it. Right now, Intel is not really leading he pack in processor terms.

No, perhaps not. However, in terms of marketing, Intel is way ahead of all other processors. The masses have been told to buy computer with "Intel Inside", they remember the crazy men in blue, and the guys in the bunny suits. Of course, they don't really know what "Intel Inside" means, but it's easy to remember and ask for. Consumers feel empowered by saying they want a computer with "500 megapixels memory, 60 googlebites hard memory, and Intel Inside".

Would it increase Mac sales by having an "Intel Inside" sticker on it? Maybe not, but it would have a lot more consumer brand recognition.

Re:Does this mean -

[Austerity Empowers]

No I think he meant to say their architecture. The PPC is a pretty well defined processor, I've used it on a number of designs way outside that of a traditional computer, but it's just a processor. You can string it to just about any impossible chain of stuff imaginable, trust me, I have. None of them were compatible with Apple's Macintosh however.

PCs and Apple's have an "architecture" defining how the chips are strung together, how expansion is expected to work, how the interrupt controller works (and yes, which interrupt certain hardwired devices are on), etc. Unlike Apples, for the PC it's not one but many standards defining their function, not one of which defines how the processor should work. Worse still, it's essentially defined as "Be backwards compatible with an PC AT from 20 years ago". No single company really owns it, although many would like to. Even the evil empire has relegated itself to "putting up with cooperation" in this regard.

I find it more likely that Apple will define their own computer architecture using Intel chips. Maybe they will do their own north bridge, in fact much of the traditional PC architecture is emulated in this device, and changing it with something else would make an incompatible system. Let's not forget that much of PC legacy crap is software as well as hardware. If Apple throws all that out the window, they may as well have defined a new Intel based system. You aren't going to install windows on it, nor will you get a regular x86 build of linux to come close to working.

Re:Does this mean -

[jdgeorge]

Exactly. Bottom line is that open architecture is superior to closed architecture. x86 is actually a shitty architecture when you get right down to it, but at least it is open.

Err... The x86 chip architecture is NOT open. If you want to produce a clone of the chip, you have to license the technology from Intel. For example, AMD has a license which allows it to produce microprocessors that are compatible with the Intel x86 CPUs.

PPC chips are at least as "open" as Intel's:

Apple, IBM, and Motorola collaborated in creating the PowerPC architecture. Apple was buying all its PowerPC parts from Motorola until they started having yield problems that prevented them from producing the high-performance parts that Apple required. freescale semiconductor/a?, Motorola's former chip division, was spun off from Motorola and is currently selling PowerPC processors for embedded applications.

Re:Does this mean -

[Daniel Dvorkin]

I would bet good money that not 10% of people who buy Macs have ANY idea what kind of processor is in the system -- in fact, the majority of them probably think it has "Intel Inside", since "that's what computers use, right?".

I think you're wrong about this, simply because buying a Mac these days is a deliberate decision. People who just want "a computer" go to Wal-Mart or Best Buy or Office Depot and get a box with Microsoft(r)(tm)(c) Windows(r)(tm)(c) and Intel(r)(tm)(c) Inside(r)(tm)(c), because "that's what computers use, right?". (Windows no less than Intel.) Using a Mac is like installing Linux on your x86 PC, in that there's usually a fair amount of thought and research that's gone into the choice.

Re:Does this mean -

[NMerriam]

"one way compatible instruction sets" means they sprinkled magic CPU pixie dust on early PowerPCs, which meant the PowerPC601 could run 68020 code if it closed its eyes and wished hard enough!

It was called "one-way" because if the PPC wished too hard, for too long, it would stay a 68020 forever! So Apple started writing software that would emulate the 68020 temporarily when the PPC was at risk.

When the Mac OS was all PPC native code (8.1? 8.5?), the one-way compatibility was turned off and all the PPC chips had to grow up for ever and ever.

The End.

Re:Does this mean -

[MarcQuadra]

Woah there, you're very wrong. The 68k compatability was done entirely in software. There was no magic dust on the early PowerPC chips. The firmware of the PowerPC line had a 68k emulator in it and the OS could supplant it with a newer emulator once it was past the boot stage.

And nobody ever turned it off, my Dad runs HyperCard 2.1, which came out a LONG time before Apple even considered PPC, on his G4 in the Classic environment, which includes the 68k emulation as part of OS 9.

As for what OS version they finally got rid of 68k code? 8.6 still had some 68k code, and I'll bet 9.2 has a few tidbits (I'd have to dig with ResEdit, and I don't feel like it now. Many of the original 'toolbox' APIs for the Mac were hand-crafted in assembler, it was really hard to take the old pascal blueprints and refactor everything for PPC, especially when you know that the Next Big Thing is going to obsolesce the toolbox anyway.

Re:Does this mean -

[EvilTwinSkippy]

(Ahem).

I'm a computer professional. A router programming, server building, cable splicing, code hacking computer professional if you care at all. In college I used to build embedded systems and I've have more than my fair share or processor architecture courses.

When I go home, I just want to turn the damn machine on and have it do what I need it to do. And that machine is an iBook. I know it has a G4 processor running a 1Ghz, the memory bus, and most of the I/O architecture. My next desktop is going to be a Mac, because my wife, a professional computer teacher who specializes in Windows, feels the same way.

Re:Does this mean -

[Lars T.]

On Intel they have SSE, MMX etc. units.

Yeah, and apart from a lot of differences, they are the same as Velocity Engine.

Re:Does this mean -

[composer777]

I'd be curious to see if the extra registers included in amd64 would be enough to speed up x86 emulation of PPC. How many registers does the PPC have?

Re:Does this mean -

[javaxman]

Please explain the process whereby Apple will convert everyone's old applications into fat binaries. Without access to the source.

If you're going to have a different CPU, clearly you're going to have different binaries. However, I'm not sure you need to be so flippant about it. Do you know about NeXTStep for Intel ? Do you know how hard it was for most companies to recompile their binaries for it? They opened up their projects, pressed "Build" in Project Builder, and it was done. That's how hard it was. The biggest problem was for folks who wrote binary data files ( the endian issue ), but that's pretty easily worked around or avoided, really. The hurdles Apple and it's third party developers would face to provide binaries for *any* gcc-supported CPU are not as huge as many folks seem to think

Yea, if you had a huge investment in current OS X software, you'd have to buy new copies ( more likely upgrades ) if you bought a different machine. That typically wouldn't be the case, though. I'm going to be using my current Macs for years. Most new Macintosh purchases ( like most new WindowsXP purchases ) require buying some new software.

However, just because it could be done doesn't mean it will be done. I just don't think they're going to start making Intel-based Macs any time soon- I can't think of a good reason. If they're in talks, it's likely to be about one of the many other types of non-CPU chips Intel makes.

Re:Does this mean -

[shotfeel]

At one time, IBM was developing a PPC that also included an x86 core. I believe it was the PPC 615.

Re:Does this mean -

[nokiator]

My gut feeling is also that the subject of Apple-Intel talks is not x86 chips. There are two other, more viable options:

1. Intel's PXA family. These are mid-rage embedded processors targeted at applications that require very low power consumption but a decent amount of compute power. Assuming that Apple is not likely to bring Newton back, the obvious target would be a video iPod kind of device.

2. To be able to run a iFlix Movie Store, Apple needs much more bandwidth to end customers. Even if this is a background service where customers maintain lists (similar to Netflix) and top few items from the list are downloaded in the background, phone and cable companies that own the last mile will not just sit back and watch Apple make money by using their precious bandwidth. At least in metropolitan areas, one way to bypass Cable and DSL providers is WiMAX. WiMax is especially suiatable for broadcasting (or multicasting) content. Intel is one of the leaders in the WiMax effort and was one of the first vendors to come up with WiMax silicon.

Re:Does this mean -

[LionMage]

In fact, I worked for Informative Graphics when there was a project underway to get a native port of their Myriad software working under NT on the DEC Alpha processor. The native port worked, but it was substantially slower than the FX!32 emulator running the x86 version, at least after running the x86 version of the app under emulation a few times. (Like the grandparent poster speculated, the emulation cached the results of opcode translation for future reuse. Eventually, almost 100% of the original application was translated and stored in a disk cache.)

Then again, the Myriad code was pretty horribly written, and optimized to only compile well under Microsoft's Visual Studio environment. I was stuck on a horrid project porting the code to HPUX and Solaris using MainWin (which basically was a Win32 implementation on top of X11 / POSIX, a porting library for lazy companies that didn't want to invest time and effort in writing truly portable code or rewriting their UI code). Granted, HP's C++ compiler sucked -- it was AT&T Cfront based, and had to be told how to instantiate templates because it had dain-bramaged template support -- but even when we got this stuff working, it wasn't very performant.

Which brings me to another thought -- if Apple switches to x86 or Itanium, we might be in for similar performance surprises. Some code will obviously benefit from a native recompile, but other code might be more performant with a caching code translation mechanism.

Re:Does this mean -

[LionMage]

This "JVMs are ass-slow" argument gets repeated a lot. There have been multiple benchmarks published demonstrating definitively that this is not the case -- in fact, Java can outperform C and C++ in many applications. Here is one reference for you to ponder. Of course, it's easier for most people to parrot hearsay rather than actually rely on empirical evidence to support their opinions.

Note that I'm not saying JVMs are inherently superior for all, or even most, applications. I'm just saying that things are a little more complicated than your one-line zinger would imply. I'm also saying that virtual machines can be very performant, with the right code-translation and caching strategies. (See my other post in another branch of this thread.)

Re:Does this mean -

[Oculus Habent]

Just this weekend I came to a realization... Microsoft just recently bought Virtual PC, and I couldn't figure out why. Now, they are offering a PowerPC-based gaming console with backward compatibility to it's x86 system.

I figure three PowerPC processors @ 3.2Ghz should be able to emulate an Xbox without troule.

Re:Does this mean -

[javaxman]

If Apple does start making machines with Intel processors, you can bet the new machines will be able to run PowerPC binaries through emulation. I don't think they're dumb enough to make all existing Mac software stop working on new machines and think people would be willing to "upgrade" their Apple hardware

Which is all part of why I don't think Apple is going to abandon the PowePC. It's too invested in Altivec, for starters. It'd be more likely that Apple would release an OS X for Intel, leveraging some existing cheap PC manufacturing, for some sort of low-low-end machine. People with newer Macs ( and PowerPC-only software ) thus wouldn't buy it, and it'd be like "look at this pathetically cheap stuff you Windows folks are forced to work with, but if you want it cheap, you can at least have it nice and OS X-y"... not that they're going to do that. But they could. By actually having some cheap-ass plant in asia somewhere cranking out nearly-standard Intel PC stuff, they could probably do it cheap enough that they'd be able to keep selling the "true" PowerPC Macs as well.

They *could*, in theory, also go the 'fat binaries' route, just like NeXT did. They stopped making their own hardware and sold _just_ software for x86 machines. According to some, it actually sold pretty well. If they did it today with OS X, it'd sell a hell of a lot better. And sure, maybe they'd throw in some sort of PearPC-style emulation engine ( maybe they'd "leverage" the existing open source project, what the heck ), that's not a bad idea, although it's clearly a stop-gap idea. I don't use Classic; people in practice upgrade the things they can and only use emulation where they have no other choice ( unless performance is not an issue at all ). People who really objected to getting new apps and moving to OS X are still clinging to OS 9's cold, dead body, and many of them will stay that way until their machines give up the ghost.

My point is that if Apple went to another CPU across the board ( not happening, folks! ), third-party developers would as a general rule, be able to offer cheap 'upgrades' as porting would be a snap, much like it was in the NeXTStep for Intel case. For most people, this would mean new software; emulation would be the exception, not the rule.

Again, I don't think that's the current strategy, unless they're finding that IBM just won't or can't get it together, in which case anything could happen. I think they'll give IBM another couple of years before doing anything drastic though, and this is just pre-WWDC media confusion. Not that I know anything... but IBM is getting a lot of dollars from these game console makers, which hopefully will fund more chip R&D and manufacturing infrastructure, eventually boosting their offerings for Apple. If not, Apple does have options.

On the other hand, the current G5 machines aren't exactly slow. They're pretty damn fast, actually. I think going Intel would have to be about price, and with the Mac Mini, I don't think that's any longer a very valid reason.

Re:Does this mean -

[RzUpAnmsCwrds]

The "Java is faster than C/C++" argument argues this"

No, it argues that a JIT compiler can make optimizations at runtime that could not be made at compile time. Which, indeed, is true. Certain specific algorithms are faster in Java than in C++.

Of course, we're talking about technicalities here. I'll agree that C is genarally faster than Java. But Java isn't necessarily slow - take the T-Mobile Sidekick, which has a quasi-Java OS and Java applications, yet maintains excellent performance on a 50MHz ARM CPU.

Re:Does this mean -

[laird]

"We will see Windows on PowerPC long before we ever see the full OS X on x86."

Ironically enough, we've already seen Windows on PowerPC and OS X on x86. Specifically, Microsoft shipped Windows NT (3.x and 4.0) and Windows CE for PowerPC, and Apple shipped (to developers) Rhapsody for x86, and currently ships WebObjects for many processors, including x86. WebObjects includes the Cocoa runtime, which means that (except for Apple's license prohibiting it) developers could ship their Cocoa applications as a single binary installer for all WebObjects platforms (MacOS X, Windows 2000+, Solaris). I've been told by friends at Apple that they still make sure that MacOS X runs on a wide range of processors (x86, PPC, etc.) in order to make sure that they don't accidentally break the portability that NeXTSTEP gave them. Of course, this doesn't mean that they'll ship it to consumers, but it's important that they keep the option open.

Re:Does this mean -

[Trillan]

No, they are not. The PPC does not support m68k code.

m68k support was done through a software emulator in the Mac OS, similiar to Virtual PC for emulating Windows on the Mac or something like MAME for emulating arcade games. The difference is that Apple did such a good job on the emulator that people like you didn't realize it existed. :)

Re:Does this mean -

[LionMage]

This is starting to get WAY off topic. However, I should point out that your point number 1 is attributable to many things, none of which are directly VM related. Poor programming is indeed to blame in some cases. You speak of "Java desktop apps," but realistically, this must be broken down into three sub-cases: AWT, Swing, and SWT, the three main GUI toolkits available to Java programmers these days. SWT has performance problems on every platform besides Windows, because SWT has only been optimized under MS Windows. AWT is no longer promoted, and seldom used, mainly because it looks clunky and doesn't provide much flexibility. Swing widgets are supposed to be 100% Pure Java, although Swing has been optimized on some platforms (notably OS X optimizes Swing drawing operations using Quartz 2D or similar, depending on version). Obviously, having GUI widgets that are written in an interpreted/JIT-compiled language is not a recipe for performance, especially for applications with rapid visual updates. But then again, a well-designed Swing app will intelligently handle redraw, something I've noticed is hard for some programmers to cope with. After all, when you can live in a Windows world and rely on hardware-acceleration to conceal the fact that your application redraws the same thing a dozen times, your programming techniques will fall flat in Swing. (I've been around long enough to see examples of GUI code in Java written by people who cut their teeth in Windows and X11, and almost invariably, profiling the code revealed cases where redraw was happening far more than it should have.)

Of course, this "desktop apps" argument is kind of a straw man argument, because these days, almost 100% of commercial Java development is on the server side, in J2EE environments like Websphere and WebLogic. Major corporations use J2EE (and JSP/Servlet engines like Tomcat) because this stuff works. It's easy to develop for, the performance is very good when using the Server VM (as opposed to the client VM, which is optimized for processes that don't run 24/7), and the toolset is very rich, which makes productivity very high.

Let's put it this way: Java wouldn't survive if it weren't performant in the server space. So obviously, for some applications, it's superior to the alternatives.

Your point number 2 is another straw man argument in disguise. The fact is that there is a lot of crap C/C++ code out there, and much of it is in libraries that are leveraged by user-written C/C++ applications. However, even ignoring this fact, you can empirically demonstrate with experiment (not so-called "intuition," which often gives incorrect answers -- ask any real scientist) that for some algorithms, a direct implementation in Java may in fact outperform a similar implementation in C/C++. The paper I cited in fact does this type of comparison -- it is an "apples to apples" comparison, implementing the same algorithms in each language and comparing the results.

Nice job on the straw man arguments, though. It's very easy to misrepresent someone else's argument and then tear it down, because you're not really addressing the other person at all; it only looks like you are.

Now, let's talk about your intuition. Most modern JVMs use what is called JIT (or Just-In-Time) compilation of the Java bytecodes. A VM that is tuned to run long-term, such as the server JVM that Sun ships, will cache the JIT-compiled bytecode for immediate retrieval and execution. As another poster pointed out, there are optimizations that can be performed at runtime by inspecting the running code's behavior; these optimizations can and do outperform compile-time optimizations, where the optimizer can only make assumptions about the run-time profile of the running code. Basically, in any case where a static compile-time optimization fails on the C/C++ side, a code implementation on the Java side will be more performant. So there's the counter-argument to your "intuition." Not that this is likely to conv

Re:Does this mean -

[Lally Singh]

Don't worry about being off topic, once the /. post leaves the front page, nobody else but us cares :-)

The Desktop App example is actually really straightforward: the delays in using Java are immediately visible to the user, and immediately comparable to C/C++ desktop apps. Delays imposed by java on the server side are less visible; as they can only be seen indirectly by a client app (and web browsers are C/C++!), and there aren't many comparable C/C++ apps on the server side.

And, btw, the document you linked to before was all scientific number-crunching. Something that wouldn't run as a server app. And I did read through the source of that, they're reasonably close implementations with each other, in a sec I'll describe why C/C++ is still going to be faster in a production environment (even with the same number-crunching code!). See next paragraph.

As for the intuition claims, what happens when you feed some profiling data back into the compiler (which the better ones accept) ? Any (good) C/C++ app development house will have automated tests running, which provide gobs of good profiling data. And again, a compiler back-end can optimize more aggressively than a JVM can, as it has high-level knowledge of the system (e.g. the source code) vs only the low-level knowledge a JVM has (bytecode only). Not to mention that the low-level implementation of the intermediate representation can be made highly compatible with the target's architecture, vs having one imposed from Sun (which, btw, was made for interpretation, not recompilation, compare with .Net's CLR which was).

And I was being nice about GC, and I'll be nice again. I'll save the debugging arguments for another day -- java programmers tend to fear problems solved in C++ quite easily. Btw, the arguments for JIT are usually similar for GC: in the "big picture" (e.g. a fantasy land where I can claim whatever I want), the overhead is nothing compared to code that we don't actually talk about.

And some more niceness: it's easier to tweak out Java code's performance: the JVM does most of the heavy lifting for you. Java's often easier to develop for (unless you use the libraries too heavily, they're mostly shit), and there are many areas where the dev effort/speed trade off favors java over C/C++. Java has a richer runtime with some nice parts in there (reflection comes to mind).

However, the claim that Java can be faster than C/C++, when good people are put on both side, is false. A C/C++ compiler simply has more knowledge of the system being compiled (the source text vs java bytecode), essentially unlimited time to optimize it (for an extreme case, check out http://www.cs.utk.edu/~rwhaley/papers/icpp05_8.ps ), and full freedom for transformations (a C/C++ compiler, knowing what you originally asked for in the source code, can generate anything it wants that fills your request; a JVM's JIT can only guess at what your source wanted from the bytecode it has to work with, limited to the strict letter of the law given by the bytecode). The paper you listed was an honest attempt at comparison, but GCC isn't optimal on many platforms, most notably intel or ppc.

This is a fair set of tradeoffs; don't let the marketing people and the kool-aide drinkers tell you different.

Dvorak

[Anonymous Coward]

"John C. Dvorak predicted this for 2004-2005."

Yes but he predicts so much crap of course he'll be right eventually.

Re:Dvorak

[tbone1]

If Dvorak predicted it, you can bet it won't happen.

The Macintosh uses an experimental pointing device called a "mouse". There is no evidence that people want to use these things.
- John C. Dvorak, SF Examiner, Feb. 1984.

Re:Dvorak

[Quiet_Desperation]

That's an even funnier quote when you consider the mouse had been invented 16 years earlier at SRI. The mouse was hardly "experimental" in 1984, and was already in use in CAD workstations. Dvorak is another one of those dumbass media figures that people inexplicably listen to. Good gig if you can get it.

Re:Dvorak

[tgibbs]

That's true. There was no evidence that people wanted to use Mice. The Mac sold quite poorly early on -- it wasn't until people wrote software that really showed what a mouse could do that it caught on.

Nonsense. The Mac came with software that showed what a mouse can do--MacWrite, MacPaint. Microsoft Excel was available almost immediately. There is very little that a mouse does on modern computers that was not demonstrated in the software included in the very first Macintosh. The Macintosh took a while to catch on because people were locked into legacy DOS software, and because people needed to be convinced that the features offered by the Macinstosh--GUI, mouse control, fully bitmapped screen with software fonts--really were so valuable that they justified the performance hit. Most other computers of the time, with low resolution screens driven by character generators were hardware-incapable of animating a cursor (there were attempts to implement mouse pointers on character mapped displays, but they were horribly ugly and jumpy and unpleasant to use).

Re:Dvorak

[gotw]

Invented at SRI, causes RSI. Heh

Re:Dvorak

[tedhiltonhead]

I thought it was invented at Xerox PARC?

Re:Dvorak

[tgibbs]

I forgot about MacDraw, but it wasn't out when Dvorak said that anyway.

MacDraw was released with the Mac; it was derived from LisaDraw.

What Dvorak said was of course trivially true, but abysmally stupid--there was of course "no evidence" at the time that people wanted to use mice, because when he wrote that statement, hardly anybody except for a handful of Xerox Star and Apple Lisa users had even touched one.

Re:Dvorak

[kevcol]

I would say that MacPaint was a demo/toy program

Sorry, pal-

I used MacPaint to create perfect reproductions of my SJSU report cards to add classes I was not actually taking in order to show I was a full time student to get various discounts (car insurance, American Express student travel program for $120 coast to coast RT flights, etc).

The fact is, MacPaint saved me money!

Dvorak

[taskforce]

If what Dvorak has predicted is about to come true, I fear the space time continuum will rupture spewing forth a hoard of evil flesh eating time daemons.

AMD?

[InsideTheAsylum]

Wouldn't using AMD be even cheaper..?

Re:AMD?

[El Cabri]

Why would it be ? You think AMD has some magical trick to make processors at a cheaper cost ? If boxed processors from AMD are cheaper than Intel it's because they HAVE TO. AMD is the challenger, and that's how it works in all domain of economics : the market leader can afford to price higher that's all. Are you privvy to the deals that are actually made between Intel and the likes of Dell or IBM ? What to you know of their high quantitiy pricing policy ?

The fact is, if Apple wants to go x86 (which I think they should), they'll consider both Intel and AMD, talk to them, see what they can get and make a strategic desicison. If Intel decides that it's worth it then they'll underbid AMD. Or maybe Apple might decide that they want a supplier that is more reliable for delivering big quantity orders given that they've themselves been bitten so often by back-ordering problems.

Nope

[fr0dicus]

ARM for iPods maybe, but otherwise, absolutely no chance. Only a fool would even think this was likely.

Re:Nope

[javaxman]

Plus I *REALLY* don't see how Apple can switch architectures at this point.

They could quite reasonably switch architectures. Or even support and produce both. This was a new reality starting with OS X, and it's strange, but it's true. Pretty much every single bit of the Apple hardware stack could run on a different gcc-supported CPU with a simple recompile. Darwin X86 is totally doable. Device drivers would be the biggest problem, but... just switching out CPUs with a slightly different motherboard and keeping everything else the same should make writing device drivers unnecessary. What wouldn't work would be very old OS 9 applications, but the vast majority of stuff would "just work" with a recompile. The structure already exists to distribute multiple-target binaries in OS X. It's been done before, with NeXTStep, it could certainly be done again.

If they were to do this they would need a damn good reason, and thats whats missing, whats the *REASON*?

Very insightful, that bit. It almost makes up of for the rest of your post... what, indeed, would be the reason ? Unless there's some sort of cost savings, or we're talking about a non-PC device, I don't see it. But it certainly could happen.

Apple hardware is expensive for two reasons, one being volume, and the other being the fact that Apple actually does R&D. But R&D is not the major factor. If they could increase volume... I'm going to guess that's going to be a difference of more like $200 between Linux on Intel and Apple on Intel, and to the average user, it'll be money well spent.

I am among those who are 100% certain that somewhere, in the bowls of 1 Infinity Loop, behind several layers of locked doors, is a PC lab with Darwin, Cocoa, and major portions of the OS X software stack running on Intel ( or AMD ) hardware. It may never see the light of day, but the simple fact that you *can* compile Darwin for X86 tells me it's there...

O really?

[yurigoul]

Stuff like this keeps coming up. Seems to be part of the Apple rumour cycle. Can we trust the source??? Using the G5 is par to of the advantage in marketing terms, as a far as i can see: think different!

Re:O really?

[Golias]

Actually, the proper English is "think differentLY."

Only if you are asking people to think in a different manner.

The Apple tag line is suggesting to people that they should think about the concept of being different.

Therefore they say, "think different."

It's not correct formal grammar, but it is perfectly valid conversational English.

Aw, jeez...

[Anonymous Coward]

...After everyone worked so hard to port Linux to the PowerPC...

Why move now?

[JabrTheHut]

Why move now? Everyone's been hearing about the dual-core PowerPC chips for months, PS 3 and Xbox 180 will be running 3-core versions of this chip, so why go Intel?

Re:Why move now?

[NextGaurd]

It would be really ironic - Microsoft's Xbox2 goes power PC but Apple goes X86? Something just doesn't feel right.

The article is quite worthless

[antifoidulus]

All it says is that "Apple will use intel chips", it doesn't state what kind of chips, but it does repeat itself over and over again. Maybe Apple will use Intel chips in an embedded device, maybe they are considering bringing back the mac/pc hybrid. There is really no "meat" to this story, but we can all speculate anyway.

Original source?

[ctr2sprt]

The WSJ reports it, but no link to the WSJ's actual story? Well, here it is.

Re:Original source?

[hab136]

The WSJ reports it, but no link to the WSJ's actual story? Well, here it is.

Requires a WSJ subscription, that's why.

Here we go again...

[tliet]

For the n-th time, what would Apple have to gain? Who would buy a Mac when they could buy a Dell. Does anyone seriously believe Microsoft would release Office for Mac OS X for Intel?

The Mac would die the day the CPU would be the same as in a generic PC. Not from a architectural standpoint, I think they could make it happen, but marketingwise.

Re:Here we go again...

[Rasta Prefect]

For the n-th time, what would Apple have to gain? Who would buy a Mac when they could buy a Dell.

Maybe someone who doesn't want his Tech support calls forwarded to Bangalore? (Not that I don't have my complaints about Apple support, but at least I could figure out what everyone was telling me, leaving out the ambiguity of figuring out whether they really sucked or whether I just thought they sucked because I couldn't figure out what the hell they were saying.)

Re:Here we go again...

[ergo98]

For the n-th time, what would Apple have to gain?

Really. I mean, everyone knows that software companies can't make any money.

Re:Here we go again...

[Momoru]

Well for one thing, anyone considering using OS X would only need to buy just the OS, not a whole bunch of expensive hardware as well. I was the biggest macophile in the past, but had to switch to using Windows for work and school...now OS X is the "in" thing, and i'd love to check it out, but I'm not going to buy a bunch of hardware just to play with it. If it were as simple as just installing a dual boot on my current PC, i'd do it in a second.

Re:Here we go again...

[squiggleslash]

Who would buy a Mac when they could buy a Dell. Does anyone seriously believe Microsoft would release Office for Mac OS X for Intel?

Hold on. This isn't the same proposal as has been made before. We're talking about an Intel-chip based Mac, not a Mac that's a PC clone. We're talking about an OS X that runs on an Intel-chip based Mac, not an OS X that runs on a generic IBM PC clone.

People would buy a Mac instead of a Dell for the same reason as they do today - a nice, well integrated, computer system with a decent OS. Few people buy a Mac thinking to themselves "Dude, this has a PowerPC! That's 32 32 bit general purpose registers, and a RISC based architecture based upon IBM's attempt to create a next generation mainframe and minicomputer platform in the late eighties!"

The only downside this would have against a PowerPC Mac is that older software wouldn't run on it. That's not an issue if you don't have any older software. About 95% of computer uses do not have older versions of any software.

And yes, Microsoft shouldn't have a problem releasing an Intel version of Office for OS X: just as long, that is, as Apple doesn't start selling OS X for generic IBM PC clones.

Re:Here we go again...

[Gilmoure]

Ya' really think Apple would release a generic OS X-X86, that would run on white box parts? Naw, they'd use Intel chip DRM to tie it into just Apple brand boxes. C'mon, they make money pushing plastic, not bits.

Re:Here we go again...

[poot_rootbeer]

Well for one thing, anyone considering using OS X would only need to buy just the OS, not a whole bunch of expensive hardware as well.

You'd have to buy OS X and all the software you'd run under OS X.

C'mon, dude. The Mac mini is only $500. Apple has put out a product that seems to address your complaints about "a whole bunch of expensive hardware" as a barrier to checking OS X out, and yet you still complain about the barrier that's no longer there.

Re:Here we go again...

[Momoru]

With a wireless card (no wires where my office is) on a 1.2ghz mac mini, the price is $578. My current computer, a dell with a 3 ghz p4, with monitor, keyboard, optical mouse etc was only $500. So the mini is still pretty expensive in comparision. But that aside, $580 might be an impulse buy for you, but $600 is an awful lot for some of us just to "try something out". And with those barebones specs of the mini, if I liked it, i would probably have to buy a "real mac" a couple of months later anyways. My point is that I already have some bucks invested in hardware...not having to scrap all that would be a serious selling point for picking up an x86 version of Tiger.

Re:Here we go again...

[DebianDog]

If the mini was a $1000 it would still be a bargain. ;-)

- Digital hub (Synergenic Applications)
- Unix based
- No worms, no viruses
- Plenty of freeware/Linux ports/3rd party apps

The Dell is a better deal if time and convenience is of "no value" to you.

Re:Moving toward standards?

[0x461FAB0BD7D2]

To get a better deal from IBM. That would lower their costs and lower the prices of the Macs.

Essentially they're shifting part of the cost burden to IBM, while keeping their share of the profits intact, in an attempt to boost their sales.

Also, it could be an attempt to make sure that IBM, with its focus on the PS3, Xbox 360 and Nintendo Revolution, does not forget about Apple.

Apple Denies

[nbharatvarma]

Some links I found some 30 mins ago in Google News

http://www.techsmec.com/index.php/2005/05/23/apple _denies_intel_rumour
http://www.pcpro.co.uk/news/73057/apple-denies-eye ing-intel-chips.html
http://www.dvhardware.net/article5037.html

Of course, one could argue that Apple wouldn't want this news to be leaked

Re:Apple Denies

[neo]

Historically when news agencies leak Apple news, they end up in court. If Apple hasn't sued anyone about this, then it's not true.

The Register says not.

[Anonymous Coward]

The Register already has an analysis on this: http://www.theregister.co.uk/2005/05/23/apple_inte l/

The conclusions are: Apple already use a lot of non PowerPC chips (iPod, AirPort base stations), so these talks may well have nothing to do with Mac's. Also, it could be a scare tactic to make IBM a bit more eager as a chip supplier.

Re:The Register says not.

[syntax]

Things may have changed from the original base stations, but back then the stations were running a 200Mhz x86 AMD.

I hate to like this idea.

[TempusMagus]

Well, for one, it would make the whole confusing use of clock speeds vs platform processor go away. It would also make it easier to emu windows software and port games. However, the new IBM PPC chips seem to kick all sorts of major ass. Why give that up? I'm betting anything this is for iPod chips.

Why cheaper!?

[jerde]

And why, now, would Intel CPUs be any cheaper?

Currently all of Intel's stuff runs hotter, so Apple would have to work significantly harder at heat dissipation issues in all but their tower designs.

And what, pray tell, do you expect them to do with little-endian issues, backwards compatibility, and all those little details?

Unless Apple thinks that neither IBM or Motorola are ever going to catch up, I just can't see them justifying the huge cost of a major architecture change like this.

- Peter

Re:Why cheaper!?

[Reality Master 101]

Currently all of Intel's stuff runs hotter, so Apple would have to work significantly harder at heat dissipation issues in all but their tower designs.

That's a design issue, not a manufacturing issue.

And what, pray tell, do you expect them to do with little-endian issues, backwards compatibility, and all those little details?

It's commonly known that Apple keeps a version of OS/X for Intel current and ready to go if they should ever have to switch because of supply problems (which has always been a real threat).

The biggest reason they don't switch is because Apple likes incompatibility where they can do it -- it locks people in. If Apple used the Intel architecture, it would be a lot easier to run their software on less expensive hardware.

Re:Why cheaper!?

[mnemonic_]

Actually, the Pentium M's run far cooler than any G4 or PPC970 processor, and Intel plans to extend the M line more into the desktop market (already some desktops using the Pentium M). The P4's probably run cooler than the PPC970's, after all, P4 mobile chips have been available for several years now, whereas PPC970's have been desktop only for 2 years. Intel's processors are cheaper because they're more massively-produced. Compare to the case of the APG-77 radar, whose price plummeted (by millions of dollars) after common circuitry in the T/R modules became used in consumer wireless ethernet gear (market expansion from .5M chips to 9M+). The 3-5% of the market Apple holds is nothing compared to the dominance x86, especially when you factor in the vicious competition between AMD and Intel that drives prices down further. Competition that does not exist in the cathedral Apple world, unfortunately.

Re:Why cheaper!?

[daviddennis]

Every company in the world supporting the Mac platform would have to recompile and reissue their software for x86. This would be a huge burden on Mac software makers.

That alone seems like a good enough reason for this not to happen short of major disaster for the PowerPC platform. And with the dual cores well on the way, according to most sources, I don't see this as a major problem.

D

Re:Why cheaper!?

[Roadkills-R-Us]

They probably wouldn't be.

But at least we'd see louder Mac Minis, because if they went with Intel CPUs, they'd need those hellaciouly large CPU fans as well.

In a Mini, that might make for a helicopter. Which would be its biggest redeeming featire - thenew, Flying Mac!

Re:Why cheaper!?

[skribble]

Provided Apple does it's job right this isn't too complex. Essentially it's a rebuild for a Cocoa App. (Next/OpenStep ran on multiple platforms). Remember... with the exception of very low level programming (i.e. some drivers, and almost no productivity apps). You write code for an OS not a hardware platform. If the OS is ported properly along with the supporting lowlevel libraries porting an app is trivial.

Re:Why cheaper!?

[Jeff DeMaagd]

I dunno, the G5s can use a LOT of power. At full load, a dual 2.0 G5 is rated for 600W. At idle, the same system still consumes 120W. That is the reason that even the G5 "FX" (PPC 750FX?) uses the liquid cooling system for the 2.5 and 2.7 models. That said, for their compute power, these PMG5s are incredibly quiet. I have very quiet Xeon workstations and my PMG5 is faster and quieter than any of them.

Probably False

[SirStanley]

This is a rumor that keeps getting recycled. It could be true this time... just like it could have been true the last 50 times it's happened.

New headline:
Erroneous Wallstreet Journal Article causes Mac Fans through out the world to riot. Killing 15

Re:Intel make chips other than CPUs

[Dominatus]

Are you serious? The G5 generates a ton of heat. Why do you think you haven't seen G5 powerbooks yet?

What really happened ...

[maxwell demon]

Steve Jobs said he liked the potato chips he was offered during an Intel presentation, and plans to sell the same chips in Apple's cafeteria as well. :-)

Predicting the future ain't what it used to be

[NMerriam]

Well, I suppose since Dvorak predicted it every year since 1988, he might well be right sooner or later. I guess that would be about the third or fourth thing he's gotten right in all that time.

Re:Intel make chips other than CPUs

[Loco3KGT]

I was led to believe we hadn't sacrificed enough virgins.

Summary of issues

[G4from128k]

Here's why this is not that likely:

1. It's just Apple trying to get better terms/service from IBM (think Dell's "talks" with AMD)
2. It will be the death of Apple's hardware division
3. Apple will have a hard time supporting the myriad boards, chipsets, and peripherals of PCs
4. Piracy/sharing (pick your preferred new-speak term) will mean a revenue-less expansion of the install base

That said, Apple's done some strange moves in the past. If PC users can just buy OS X86 for $99, they might give Mac a try. It wouldn't take that high conversion rate for OS software profits to easily replace hardware profits. I'd bet that Apple makes nearly as much profit on a sale of Tiger as it does on the sale of it slower-end machines.

Re:Summary of issues

[Dun Malg]

# It will be the death of Apple's hardware division # Apple will have a hard time supporting the myriad boards, chipsets, and peripherals of PCs # Piracy/sharing (pick your preferred new-speak term) will mean a revenue-less expansion of the install base

Why is there always the presumption that a system with an x86 CPU will be PC compatible? Someone postulates that Apple may be considering using Intel CPUs, and everyone makes the giant leap of (il)logic that Apple is considering adopting the PC platform as a whole. It is entirely possible to use the same CPU in a totally incompatible system. Look at the Original 68000 based Macs. Were they compatible with the 68000 based Amiga? It is only logical to assume that were Apple to dump IBM and adopt (say) the Pentium-M as their new CPU it would be installed on a proprietary Apple motherboard, not a $40 Abit or Tyan from Taiwan.

Intel makes a lot more than CPUs

[stratjakt]

Flash memory, network controllers, raid controllers, memory controllers, etc and so on.

Maybe they just plan to use an intel PIC to control the little blinky power light. Or one of Intel's DSP's to make the iPod not sound like a 70's era 8-track.

It's highly unlikely this means OS/X for the PC. Apple would never give up their fiefdom.

unbelievable

[cg0def]

It is amazing how many people still believe that PPC is vastly superior to x86. (and yes it is x86 and NOT 386). What exactly do you think a PPC can do that an x86 cannot? It is not that hard imagining Mac switching over to silicon provided from Intel. The IBM CPUs cost them a bundle and plus Intel has the capacity to produce huge amounts of CPUs while IBM doesn't. Also IBM being so huge and going in so many directions does nothing to promote the CPUs that they make. On the other hand Intel does a great deal of advertising and this will help Mac a lot. Not to mention that an x86 compatible MacOS X will brings a boat load of new customerts for Apple and probably double the value of their shares if not even tripple them. So moving over to x86 especially now that x86 is moving towards multiple core CPUs is a great idea and I am all the way behind Apple if this turns out to be more than a rumor. Hey I'll be one of the first to buy the x86 version of MacOS X.

Re:unbelievable

[Kjella]

It is amazing how many people still believe that PPC is vastly superior to x86.

I don't. But it is considerably much more "on par" than the G4 ever was, and they weren't changing then, why would they now? It is almost impossible to emulate PPC on x86. I don't know if x86-64 is any better, it should be because of the added registers but it'd probably still run slow.

I'm sure apple has a drop-in ready box for running x86/Mac, but it'll essentially be the same box as today except for the CPU socket, at the same price (no, the CPU is a damn little part of the total box). How would that bring any new customers at all? Not to mention x86 OS X is so much easier to pirate than PPC OS X. Chances are if you are running OS X, you probably paid for it at least once every HW upgrade.

Kjella

Re:unbelievable

[AKAImBatman]

It is amazing how many people still believe that PPC is vastly superior to x86.

I'm amazed at how many people still think that any performance gap (real or perceived) actually matters. The majority of your PC's performance now comes from the size of the bus, the transfer rate of your disks, and how much memory you have. No one really *needs* a 5GHz processor to run a wordprocessor, email client, MP3 player, or even something more intensive like a graphics editor, video editor, or sound studio. Even games now rely far more heavily on the GPU than they do the CPU.

Re:unbelievable

[AKAImBatman]

Good point, but Apple still ships G4 chips with an obsolete 20th century-style bus.

Apple ships G4 chips with low-end hardware (where price > performance) and laptops (where low heat & battery > performance). The PowerMac line has the desktop performance demons that are strong competitors to the latest Wintel offerings.

If PowerPC was truely competitive, Apple would have stopped shipping G4 chips a long time ago.

If Intel/x86 was truely competitive, they would have stopped shipping Celeron and Pentium M a long time ago.

Intel has very good marketing. That's why people think that they're always fastest, even when they're not. Low-end and Laptop PCs suffer from many of the same design aspects as low end Macs. This is an intentional characteristic, as it allows for machines to be cheaper for your average user. The high end user is going to have to pay fairly heavily for a top of the line machine.

As for laptops, Apple has recognized that these machines are not designed to be high-performance machines, but rather computers designed for doing work on the road. So they've designed them to have long battery lives and low heat disappation. Up until the introduction of the Pentium M/Centrino design, x86 laptops tended toward high performance, short battery lives, and massive heat dissapation. (Trust me, it's quite annoying when your Dell is dead after an hour and a half, and the guy with the Mac next to you decides to stop working and throw in a DVD for the next two hours or so.)

Re:unbelievable

[AKAImBatman]

That's a bit of a blanket statement...

No, it's not.

As someone who lives in Photoshop, InDesign and Illustrator for 45+ hours a week, I wouldn't complain if my filters were twice as fast. Not every cpu-intensive process is bus- or disk-limited.

Your photoshop filter probably *is* bus and memory limited. For one, your processor needs to move the memory in and out of the processor quickly. If the bus is saturated, then it can't feed the CPU any more than it already has. Which means that the CPU can't run any faster than what the bus can feed it. And if you run out of memory, you'll have to start accessing the disk - which is several magnitudes slower than memory. Improved disk performance (e.g. Serial ATA) can help hide some of that swapping.

I regularly peg my dual 2ghz g5s. If it were dual 4ghz, I would peg that too, just not as often.

Pegging a CPU is not the same as getting 100% of the CPU's performance. If the bus can't feed the CPU, it's going to start running wait states which look like normal CPU usage to the OS. Most people would be amazed to know that the time spent in wait states can easily be half or more of a CPU's processing time. i.e. It's all about how efficiently the bus can service the CPU.

Trust me. Every time you get a new machine, most of the performance increase comes from systems other than the CPU.

Re:Thoughts on this...

[dick johnson]

The codebase for the OS is already done. Nextstep, currently known as OS X, has always run on Intel chips.

As for the developer base, it's a trivial matter to port native (cocoa, not carbon) applications to Intel from PowerPC

You would need to recompile it. But that is not a huge matter for someone who has programmed their application in cocoa.

Re:rumor?

[Batzerto]

This would be GREAT news for Microsoft. No more whining that they are a monopoly that leaves consumers nowhere else to turn.

Re:rumor?

[cgenman]

The WSJ does have an excellent reputation, but remember what it says... "Chips." Nowhere does it say x86. This could be an agreement for Intel to get into the PPC business, which would be a great supplier coup for Apple, or it could be an agreement to switch to cheaper Intel wireless networking chips. Maybe Intel will build Apple's ROMs. There are a lot more chips in a computer than the main processor, and nowhere does it say they're thinking about switching suppliers for that or the base architecture for that.

And maybe they won't be used at all. The WSJ says they are in talks that "could" lead to using Intel chips. It's known that at least one version of Apple's OS was up and running on an x86 chip, in the same way that Microsoft had Windows up and running on a PPC architecture. It's also known that Apple talks a lot.

I'd say the chance of a complete platform shift is slight, as backwards compatibility from x86 to PPC would be a nightmare. But Intel supplying PPC chips to Apple, after the years of languishing Apple went through before IBM could deliver a G5? That's a lot more likely.

Never Happen

[kaltekar]

'Analysts' have been predicting this for years, saying its cheaper, more flexable, etc. Apple has too buch invested PPC, if Apple switched to x86 EVERY program on the current platform that is optimized for PPC would have to be reworked to run on x86. That and Apple has too much of a proformace advantage with t he G5.

Re:Intel make chips other than CPUs

[pizen]

Apple doesn't want to lose it's userbase.

ba-ZING!

Weird timing

[Per Abrahamsen]

With IBM CPU's powering both the new XBox and Playstation, one would imagine that volume production for cheap Mac's would be possible. Is there any reason you couldn't use a XBox 360 CPU in a Mac?

Cell

[mikeee]

I wouldn't think so. You could, but Cell is a weird chip; not so much like a G6 as a G4 with 7 (!) super-VMX coprocessors.

With proper coding, performance for something like Photoshop might be fabulous, but I doubt it would be efficient for general-purpose computing.

Re:Intel make chips other than CPUs

[TylerL82]

I'm sure that Apple could make a 2.5" thick MegaBook to compete with the high-end desktop-chip-in-laptop offerings of Dell n' Friends.

Is that what everybody wants from Apple?
An unwieldy leviathan with an hour battery life?

It's not about CPUs

[CptSkippy]

Who said this had anything to do with CPUs?

Intel = Flash Memory God

iShuffle = Flash Memory MP3 Player

Apple Already Uses Intel-Intel Uses What Was Apple

[adzoox]

Apple no longer owns a stake in ARM (sold it all to Intel), but they do have experience and could easily hire/rehire programmers that worked on Apple/ARM devices (read as Newton)

My guess is that this will be for a new ARM processor for the iPod - the custom chip is probably too weak for advanced features.

Intel has been wanting to move to RISC chips for some time - maybe they want to become a CELL/PowerPC production partner to catch on the wave - 100% of the gaming world will be using PowerPC or PowerPC deritives in the next year - year and a half -

Intel now owns the largest stake in ARM (bought from Apple) - this is the processor in the majority of PocketPCs, Palms, and GPS units. So - this Intel processor is most likely for a new device or even the iPod.

Also take into account that USB2.0 chipsets are currently made by Intel (and others) and that Apple uses Intel chips in the XServe line for RAID I/O.

Re:Intel make chips other than CPUs

[TylerL82]

I mean, honestly, what about this graphic says "fairly cool" to you?

The fact that those are large fans that aren't spinning at full speed.
They keep the computer as cool as it needs to be while being much quieter than the 2-3 fan PCs with fans spinning their lil' hearts (motors?) out.

Dvorak

[jusdisgi]

PCMag's John C. Dvorak predicted this...

Even a stopped clock is right twice a day...

In all seriousness, though....that article is a good chuckle. He was calling for Apple to switch to Itanium for Christ's sake, and then license OS X for non-Apple hardware with a Windows compatibility layer. Bwahahaha! Come now, John, how do you dream this shit up?

Equally hairbrained is his theory of a dual-processor PowerPC/Itanium machine; he seems to have the idea that they could just duck-tape two motherboards together and have it pick and choose which processor to run what executable on...clearly, he's using drugs.

If Apple does this (which I am extremely skeptical of at this point) I expect it will be at the very lowest end....completely backwards from what Dvorak predicted. The reason is simple: IBM's Power architecture is plenty fast, and they aren't going to get a performance improvement by switching to Intel. The only benefit available from such a switch is cost.

Re:Intel make chips other than CPUs

[imroy]

Intel bought the StrongARM processor design from Digital a number of years ago. They now produce them under the Xscale brand. They've been used in heaps of devices, including the Compaq iPAQ, and lots of small embedded boards. Apple has previously used AMD's MIPS-based processors in some of their Airport AP's. Given the Xscale's low power/heat and relative processing power, I wouldn't be surprised to see Apple use the Xscale in another funky little portable device.

Straw man, meet the overrated rumor troll

[ianscot]

I don't see any "PPC is vastly superior posts" -- other than yours, which refutes this seeming non-starter of an argument by scattering a bunch of the usual rumor rationales in the hopes that some of the seeds will take root.

Meanwhile none of your points address the obvious problems pointed out elsewhere: that Apple has denied the rumor (for the umptieth time), that Apple's a hardware shop that might not compete with Dells that could run the same OS, and so on. Breathless arguments on the level of "probably double the value of their shares if not even tripple (sic) them" are even less convincing than a John Dvorak prediction, which is pretty far down the food chain for me.

If this is true, it means Apple has decided to go head to head with MS as an OS shop and make its money that way. That's what people call a "paradigm shift," and a radical one, at a moment when every recent example shows us that Jobs is continuing to do what he does best: sell the hardware by using the OS and interface grace points as a break-even loss leader. (iTunes as loss leader for iPods, iLife suite to sell each new generation of boxes, and so on.)

Re:Apple Already Uses Intel-Intel Uses What Was Ap

[Phil John]

100% of the gaming world will be using PowerPC or PowerPC deritives in the next year - year and a half -

You couldn't be more wrong. 100% of the next-gen console gaming world will be in the next year and a half, however, everyone who plays handhelds (nintendo ds, gameboy advance, sony psp etc) and all us PC gamers (of which there are considerable numbers) will still be using other chips besides PowerPC/CELL

Re:Only one thing left to do!

[GaryPatterson]

You sicko!

Why do you want to tape a duck? Is that some sick animal bondage fetish you have? Some sort of Beatty-esque perversion, but with less hamsters?

Or... did you mean *duct* tape?

No... the animal bondage thing seems a simpler and more believable explanation.

Re:Apple Already Uses Intel-Intel Uses What Was Ap

[stevel]

Intel now owns the largest stake in ARM (bought from Apple)

Sigh - how soon they forget. Intel's ARM technology was acquired from DEC, not Apple. It was DEC's StrongARM that was "bought" by Intel as part of the settlement of the patent infringement lawsuit brought by DEC. Not just the rights to the processor and the architecture license, but the Hudson, MA chip fab that made the processors.

As far as I know, Apple has had no involvement in ARM.

Move Along. Nothing to see here.

[Frobozz0]

*ahem* bullsh*t.

There's a saying about hell freezing over and _something_ to do with Steve Jobs using Intel CPU's.... hmmmm.

I think these people misinterpreted the evidence for CPU's. Apple uses Intel chips in their computers NOW... just not CPU's. And for good reason:

1) Intel chips are NOT cheaper. Any difference is negligable.
2) They don't run faster (AMD keeps pace, but not Intel.)
3) They'd have to recompile every app made for one architecture to run on another.
4) They run hotter.
5) Steve doesn't like Intel CPU's.
6) Steve doesn't want to piss off Microsoft by being THAT agressive in their turf.

Itainium

[essinger]

Is it possible that Intel has found someone who wants to buy their Itanium cpus?

Re:Why doesn't apple use the Playstation3 chips?

[Utopia]

PS3 CPU makes sense when you have lots of floating point operations.
It isn't any faster when it comes to general purpose functions.
Plus By the time it goes into mass production next year the standard CPUs will be much faster.

Re:Apple Already Uses Intel-Intel Uses What Was Ap

[adzoox]

Apple Sells ARM shares after discontinuing Newton, upon Steve Jobs return to Apple

Apple WAS a MAJOR developmental partner in ARM for the Newton - the processor line was GREATLY enhanced during the Newton run.

So was Sharp - probably shoulda googled before that statement!

John Dvorak predicted it?

[atomm1024]

Yeah, big deal. He predicts that every year. Maybe he was lucky this time. :)

Not for Macs....

[SJ]

XServe RAID already uses an Intel IO chip.

Airport Base Stations use (or at least they used to use) a 486.

iPod probably has (or will have) some sort of ARM chip in it.

The XNU Kernel has the ability to assign certain types of tasks to certain types of CPU. There is no reason why a Mac could not use both a PPC and an x86 in the same box.

Intel make kick-arse network chips.

Who said anything about these going into a Mac? (New product?)

This could mean several things

[Eccles]

Intel could be arranging to supply Apple with non-CPU chips.

Intel could conceivably be arranging to be a source for G5 or G5-compatible chips. As one of the world's largest chip manufacturers, they've got lots of resources and access to technology. It's more central to their business than chips are to IBM.

The OS X on x86 path seems the least likely to me. App writers don't want to support what is essentially another platform, nor does Apple.

Intel are more than just processors

[el_womble]

I can't imagine why Apple would want to move towards x86 hardware, but there are many reasons why I can see Apple and Intel having a lot to talk about.

• Intel make a lot of chips. Apple / IBM in comparison do not, but that doesn't mean that Apple doesn't want to. Intel could become a licensed manufacturer, and pick up the slack if volumes get too much for IBM to handle (in the wake of PS3 and XBox 360s).
• Intel know a lot about 90nm technology. They have several patents that would no doubt make IBMs life a lot easier when it comes to making a G5 that works in a laptop (without sterilizing the user) and pushing the G5 beyond the 3GHz barrier
• Intel make other technologies that Apple would be interested in, WiMax being the most obvious
• Intel, have the potential to be great innovators. They're reaching the limits of what they can achieve with x86 because Microsoft are unlikely to want to support a new architecture anytime soon. Apple could offer them an oppertunity to try something new, and maybe make the next big thing in processors (if they don't already have it up their sleves).
• I could even imagine a 'G6' or similar with a x86 instruction decoder. We all know that x86 instructions are internally reduced to RISC like microcode, why not bolt one onto the the front of a G5 and remove the software emulation in virtual PC? (ok, this is scraping the barrel)
• Intel inside sells 200 million units a year, maybe that badge could make a difference to Apple sales - even if they used a different instruction set

Re:Fat Binaries my not solve everything

[fitten]

Technically, those PPCs that can flip endianness on the fly are actually native in either mode. Set the mode and until you change it back, the PPC *is* a little endian CPU (or big endian, depending on how you set it).

Data storage is an issue but not one that is that complicated. In fact, it's not that painful to do endian swaps on x86-64 processors, for example, because there's a dedicated instruction in the ISA to do it. If the binary data files have an identifier in them (version number, etc.) then the swap can be done on the fly pretty fast and easily. I worked on a product (back in the early 90s) that had to have all I/O (network, HDD, etc) able to handle endianness issues on the fly because our product ran on both big- and little-endian machines and all were expected to work together in any combinations (server side on either type independently of client side being either type). It's not that hard if your software is written reasonably well (all particular I/O handled in libraries inside your code instead of spread out all over Hell's 40 acres).

And I agree with you about XML being a pig... it uses a lot of CPU cycles but at least it's a memory hog, too /rolleyes

Re:rumor?

[timster]

I'm certainly wondering about the prospects for an Intel PowerPC chip. Maybe the CPU geeks out there can tell me how possible it would be for Intel to develop a microcode (or other relatively simple modification) that would allow for a version of the Pentium-M that ran a PowerPC instruction set. Possibly this could be the chip for the next Powerbook, if indeed the G5 turns out to be too big and too hot to make it into a portable, like, ever. A PowerPentium, as unholy as it sounds, could be a great notebook chip.

Dvorak

[Jeremiah Cornelius]

Self-cancelling statement: "Interestingly, Dvorak predicted..."

NOPE

[kaiwai]

I'd say the most likely scenario, as outlined by some:

1) ScaleX (ARM) based processor for the iPod, possibly to be able to power the H264 videos that they'll be selling soon - music videos first, then work up to movies once the studios feel comfortable with the idea - I'm sure Steve will win them over with his charm.

2) Another supplier for the wireless chipset. Currently Broadcom supplies the wireless chipset for Apples Air Port Extreme; coupled with the use of Intel NICs, and the move to try and lower costs with the mini-Mac, you might just find that Apple might try to negotiate a better deal if the computer chipset + wireless + nic were all Intel.

Mod parent wrong

[calambrac]

This is false! Kids, listen: If you want to get laid, buy a Powerbook, get some black-rimmed glasses and a dog-eared copy of a Thomas Pynchon novel, and go find a good coffeehouse near a university. Grab a table near a napkin dispenser. Do not open the Powerbook but place it conspicuously on the table in front of you. Pretend to read the novel. Make eye contact with the grad student across the way and smile.

If things go well, she will decide that she needs some napkins, and while gathering them together will accidentally drop some on the floor. Help her pick up the excess paper and make a stupid little joke, something like "Oops, there go some trees." She will then say something like "I love Pynchon" at which point you reply "Have you seen Zak Smith's illustrations for Gravity's Rainbow?" You will then open the Powerbook and visit the site via a bookmark in a folder named 'Diversions'. It is important that she not see the folder marked 'Linux stuuf' or 'pron'. Spend the next thirty minutes saying things like "I really do think media is ultimately the message" etc. If you successfully complete this sequence of steps, sex is all but guaranteed.

Re:Mod parent wrong

[Gauchito]

How that will actually go with a Slashdotter:

1) Grad student notices that book is held upside down and slashdotter's eyes are staring unblinkly at student's breasts, as in a vain attempt to use the Force to rip open shirt and unhook bra (act that said slashdotter is incapable of with two friends helping and a 200 page instruction book).

2) Mac dies a short-circuiting death after being drowned in 5 inches of said slashdotter's drool, and in an ironic twist wipes 5 years of carefully and selectively collected porn.

3) Ducks are seen swimming in coffehouse's newly created indoor lake.

Re:Mod parent wrong

[freeweed]

Man.

A hooker is every so much cheaper than a Powerbook.

Plus, they don't care if you never call them again.

Additional bonus in not having to be nauseatingly pretentious.

Re:Mod parent wrong

[radish]

I don't know what's funnier - the post itself, or the fact it has a +1 Informative. :)

Re:Mod parent wrong

[dcam]

On no circumstances actually open the Pynchon novel. You will be scarred for life.

I made the mistake of reading Gravity's rainbow on the recommendation of an (at the time) girlfriend. In my entire life I do not believe I have come across a more useless and unpleasant way to spend my time.

Re:Mod parent wrong

[brwski]

Do not listen to this advice!

Gravity's Rainbow is well worth reading. It does, however, require that you pay attention. There is no hand-holding in this novel. Give it a chance --- this book is *funny*, while also being a great WWII novel.

500 bucks

[5n3ak3rp1mp]

500 bucks, about as much as buying Windows XP Pro, for the Mac Mini, will get you a seat. Set up a KVM switch and get the OS X-perience ;)

You WILL like it, and find a use for it, even if it doesn't become your primary machine.

Disclaimer: I use a Windows laptop for work and have a beloved G5 at home.

Intel fabbed PPC chips?

[parvenu74]

Doesn't Intel have all kinds of chip fabrication capacity that could -- in theory -- be converted over to stamping out PPC chips instead of x86 chips? IBM would need to license such a move or sell the IP, but is that such a reach considering that IBM has been spinning off their hardware businesses (Hitachi hard drives, Lenovo Desktops) of late?

And if Intel were going to produce a CPU for Apple, why is the assumption x86? Mac OS X Server on Itanium, anyone?

Mostly likely, though, is that Intel's wireless chips are on the menu. WiMax is around the corner and we all know how Apple oh-so-loves to be on the bleeding edge of technology...

Re:OMG!

[nullhero]

What are the chances of seeing an x86 port of OSX??

There is it's called OpenDarwin. It runs on the x86 the only thing is what make Mac OS X a Mac OS X is it's window server and os libraries - core foundation, cocoa, etc. Check out the site for Darwin: http://developer.apple.com/darwin/. They have a link to the x86 port.

Re:Low end Itanium possible - Apple software stack

[AJWM]

It's not really that crazy. I hypothesized the Itanium idea in another message before I'd read the parent.

The code base isn't that much of an issue. At the base, it's Unix, and porting Unix-based apps from one architecture to another is relatively painless. (Endianness is an issue, but the IA-64 can handle either big- or little-endian data (instructions are always little-endian)). Of course, a PPC emulator would be required for migration, more easily done on IA-64 than x86[-64].

The Itanium had speed issues, but this has been addressed with the Itanium2. As far as floating point goes, the IA64 has always been a screamer. Virtualizing an x86 PC would be a breeze.

There are still plenty of other explanations for Apple and Intel to be talking (if indeed they are) which make more sense, explanations that have nothing to do with Mac CPUs, but the Itanium idea isn't that whacky. Going to the PPC in the first place was pretty much an inspired leap, Apple may be gearing up for another one.

Re:Apple Already Uses Intel-Intel Uses What Was Ap

[norwoodites]

Apple uses ARM in the iPod already and hired a person to work on GCC for ARM. (this was all based on public knowledge).

PPC not an edge / x86 Mac not a PC clone

[AHumbleOpinion]

Not to mention, the PowerPC processor is the only edge Macs have left on PC hardware.

Not true. Apple has two edges. (1) Complete control of hardware and operating system. (2) Mac OS X. Neither of these are PowerPC dependent.

If Apple goes x86 the Mac will simply be an overpriced PC running a pretty gui on top of BSD.

Also not true. An x86 Mac would not be yet another PC clone. Apple could keep much of its current proprietary design and ignore IBM PC compatibility. A computer's architecture is much more than it's CPU.

Re:PPC not an edge / x86 Mac not a PC clone

[AHumbleOpinion]

x86 processor (if they change)
standard pci (and AGP) bus
commodity ram
ide/sata drives and connections
All the hardware interfaces would be 100% compatible with a pc.

You are not getting low level enough. Think timers, interrupt controllers, etc. All this low-level baggage being carried around from 8088 days. Now toss in the various glue logic ...

In other words you are only referring to things that plug into the motherboard, I am referring to how things are implemented on the motherboard.

Apple possibly fed up with IBM internal PowerPlay(

[Been on TV]

Having worked in Apple product management and been recruited to and used to work for IBM at a time when they wanted to put Mac OS on IBM PPC hardware (gosh, that's got to be a decade ago...), I would think that the article in the Post is a sign that Jobs has just about had it with IBM internal politics.

There are parts of IBM that do not give didley about Apple - actually a lot of IBM-ers talk about Apple as they wished it off the surface of the earth. There are of course folks in Microelectronics and some Linux on Power guys who care, but the rest...

If IBM really cared about getting more PPC based systems into the market, they'd have IBM Software make sure Apple was properly supported both on the client side, but also on XServe with their server software products. You don't see much of that.

The PowerPlay(TM) that is going on inside of IBM, and what is probably seriously hampering Apple these days, is that IBM is trying by all means to protect its high-end server business. In which the POWER processors (and dual core) play an all important role both in the iSeries (former AS/400) and pSeries (former RS/6000). These are low volume, very high margin products that sustain two ecosystems in IBM with revenues and margins that far exceeds any business IBM will do with Apple this century.

With Apple eagerly wanting to use dual-core PPC chips in, not only dual processor systems (customers chairing on the side), but possibly bringing both 4 and 8 processor systems - both workstations and servers, to the market, IBM's Enterprise Division will increasingly see this as a threat to the i and pSeries servers. Apple will, with a completely different price-point on servers in particular, significantly threaten to alter the margins IBM has on the low-end to midrange i and pSeries systems.

IBM got a very rude awakening seeing Apple XServe hardware finding the way into some of the worlds fastest supercomputer configurations at a fraction of the cost of then priced IBM hardware.

Now, with a possible 4 and 8 processor XServe out the doors, the rocking of IBM's boat would still continue. Why? Well, IBM is to a larger and larger degree touting both iSeries and pSeries ability to run Linux software both natively in AIX and OS/400, but also in logical partitions, as one of its major features and selling points. Guess what? Apple can run Linux software too.

The relative ease by which Linux software can be made to run natively under Mac OS X combined with much lower priced hardware, will make IBM's iSeries and pSeries customers increasingly ask why not to switch if all they want is the ability to run Linux software on PPC.
Such a scenario could put tremendous strain on the Enterprise Division's margins. Which is why there are forces internally in IBM who do not want Apple to have the powerful PPC chips Steve Jobs needs to transform Apple into a success in the enterprise market. They probably try to put all kinds of restrictions on what systems he can build with those chips, if he gets them.

Intel does not play these games. Which is why a processor switch may be attractive for Jobs.

Of course there are all kinds of problems with the existing installed base in terms of binary compatibility of software, but they have lived through this before without too many problems. Apple knows how to handle a processor switch from before and I think the OS will handle another chip excellently given the long time Apple has had to prepare for this.

Now for the market? As another guy so excellently put it in a post; 95% of the market does not have the problem of binary compatibility of software under Mac OS X.

Re:rumor?

[sld126]

It's also well known that Apple already uses AMD chips:
http://www.vonwentzel.net/ABS/Evolution/index.html

Doesn't it seem more reasonable that they're upgrading their wireless systems from AMD to Intel, than fundamentally changing their core machines/processes/software? I think so.

Facts (not) interfering with the fun

[mihalis]

I contacted a reporter who filed this story for one of the on-line financial news websites. He confirmed the WSJ did actually say Intel CPUs for Apple PCs.

Yellow Box

[toph42]

I noticed something profound: QuickTime 7 is built on Cocoa (that's why it is Mac-only right now). In order to run on Windows, Apple will have to revive the "Yellow Box" (the Cocoa frameworks on Windows), at least in part. I'm hoping that this means they are about to re-release the Yellow Box with full updates. That would immediately and immensely increase the market for Cocoa apps. And if the Cocoa frameworks are simply included in the QuickTime 7 install, then Cocoa apps compiled to x86 for the Yellow Box could simply list QuickTime 7 as a system requirement. :)

Macs on x86 != standalone OS X for x86

[saddino]

As usual per rumors of this type (although I suppose this source is somewhat more credible than Dvorak's wild guesses), people assume that such a move (Mac hardware based around x86) means that Apple would sell a standalone version of OS X that runs on any x86 hardware. My guess: not likely.

Apple would still be selling a closed-box solution and thus provide drivers for their platform only. OS X would still run on Macintoshes only, notwithstanding their x86 internals. A custom chipset similar to the old MacOS ROM could also help prevent tinkerers from trying to "extend" OS X to non-Apple machines. As others have mentioned, Apple doesn't want to get into the "support every x86 platform" game..."it just works" is still a motto for them.

Oh, for Pete's sake..

[jcr]

This same rumor has popped up just about every year since 1984. Dvorak has been predicting that Apple will go x86 every time he wanted to troll for ad hits.

Apple buys a bunch of parts from many companies, including Intel. If Apple execs are meeting with intel execs, its' just as likely to be about ethernet controllers or wi-fi transceivers.

-jcr

why Apple would be PC-compatible or bust

[G4from128k]

Why is there always the presumption that a system with an x86 CPU will be PC compatible?

I agree with you that Apple does not necessarily have to go PC compatible. But there are two very strong reasons that Apple would only buy Intel CPUs as part of a conversion to PC-compatible hardware.

First, Apple has always suffered from the "its more expensive" criticism (some, including I, disagree that Apple is more expensive, but that's beside the point of public perception). Adding an Intel chip will do nothing for Apple's costs (may even increase them). Only if Apple goes PC compatible will it gain access to ultra-low prices associated with Intel hardware. Apple will never beat Dell at the low-cost PC game, but could co-opt Dell's economies of scales by releasing OS X for x86.

Second, Apple gains access to a massive install base of PCs. Apple sells only 3 million machines per year versus worldwide PC production of 200 million machines per year. I'd bet there are more than half a billion PCs that have the heft needed to run Tiger (from what I have seen, Tiger can run on a 400 MHz G3). If just 3% of those PCs converted, Apple sells 15 million copies of Tiger immediately and another 6 million copies per year afterward. This, alone, would triple Apple's marketshare. Given the high gross margins on software (vs hardware), Apple could afford to lose some hardware sales. But this is only possible if Apple sells a true PC-compatible version of OS X.

Re:Apple Already Uses Intel-Intel Uses What Was Ap

[wkcole]

As far as I know, Apple has had no involvement in ARM.

As you appear to be completely ignorant of ARM's origins, why bother making such a statement?

See http://www.arm.com/aboutarm/milestones.html and scroll down to where ARM describes their origin as an independent company. ARM was initially a joint venture of Apple, Acorn, and VLSI. Selling off their shares of ARM was part of what kept Apple alive in the late 90's.

article misquoted

[eqkivaro]

Wall Street Journal reports that Apple will agree to use Intel chips

What the article really reports:

Apple Computer Inc. has been in talks that could lead to a decision soon to use Intel Corp. chips in its Macintosh computer line, the Wall Street Journal reported on Monday

Apple hasn't agreed to anything.
-c

Where is AsSeenOnTV?

[IdahoEv]

ASOT is strangely silent in this thread.

Punditry

[hey!]

The chief measure of successful punditry is not accuracy, but credibility. Credibility is not based on any particular insight on the part of the public, but on three factors:

(1) Telling people what they already know or are being told by other credible sources.
(2) Being considered a credible source.
(3) Thinking of arguments that sound good suporting what everyone thinks is going to happen.

If there is a bit of recursion going on here, it's simply because the basis of credibility is so flimsy. It also means that credibility is self-reinforcing, which means the hardest thing about being credible is getting on the credibilty gravy train. Which is good, because there are limited spots avaiable.

Mr. Dvorak used a time honored method for obtaining credibility of getting in early, on the ground floor.

From a technical perspective he's a bit late on the Mac/x86 speculation though, which has been rife for nearly twenty years now. However, this is actually a highly sophisticated bit of punditry timing. Apple had been off the punditry radar screen for nearly a decade at the time. You simly cannot excercise punditry on something nobody else is thinking about -- novel ideas have no basis for sounding credibile (see above).

However, by 2004, it was apparent that Apple was no longer irrelevant, that it had not only stopped the bleeding but had built a successful business, established valuable and powerful brand identity, and had reasserted its influence as a design leader, not only in the computer field, but beyond. So people started thinking about Apple again. And, in the same way that old English roads still bear the ruts of Roman chariots, their thoughts naturally fell into the grassy ruts of the MaxOS x86 idea.

Mr. Dvorak's 2004 prediction bears the hallmarks of expert punditry. First the conclusion is public property so well broken-in that nobody is apt to mind if it takes a bit of additional abuse. Secondly , of course, is the exquisite timing that only an ear planted firmly on the ground of public opinion can execute, falling on the heels of Apple's successful iMac by a mere six years. This is probably, ifyou will permit me a bit of nelogizing, the minimal period needed for effective punditric credibilogenesis. Any shorter and you're talking about something that nobody is thinking about yet -- disaster. Any longer and all the good theories for what everyone expects to happen will have been taken, and the whole idea will have to be put back on the shelf for five or more years.

To the dark Side...

[deviantphil]

Oh no....apple is going to the dark side.

FX!32

[falconwolf]

Digital did this with their FX!32 emulator. They could run Windows binaries at pretty decent speeds on NT Alpha

Not even close. I got an Alpha several years ago setup as a dualboot with NT and Linux, but I've hardly even used NT because I wasn't able to install that many apps even using FX!32. The only commercial software I was able to install was Borland C++ Powerbuilder. I got the Alpha because it was said you could install almost any app that ran on NT but came to find out later only "well behaved" apps will install. Now what "well behaved" means I don't know.

Falcon

hmm..

[Cryptnotic]

You weren't paying attention either. NP Complete problems are definately not unsolvable algorithmically. They are problems for which there exists a solution solvable by a non-deterministic machine (a fictional machine which can be in any number of states at a time) in a polynomial amount of time (i.e., O(N^i), where i is some constant integer). NP Complete problems are also such that they can be mapped onto other problems in NP Complete. The problem with these though is that they usually require exponential or even factorial time and/or space in order to solve them. Sometimes these solutions are "try every possible solution, pick the best one". It's a definate algorithm, but as the problem size grows, it will be impractical.

The fundamental question in computability theory is "Does P equal NP?". (Recall that P is the set of problems which can be solved by a deterministic machine (more like a real machine) in polynomial time). NP-Complete is interesting, because if a mapping were found from a P problem to an NP-Complete problem, all of the problems in NP-Complete would be solvable in polynomial time on real machines. Last time I checked, the answer to this question was unknown (no one has been able to proove that P and NP are different). Obviously, P is a subset of NP, but is it a strict subset or could the two be equal?

So, anyway, P and NP both consist of algorithmically solvable problems. For an example of an unsovable problem, look up the "halting problem". The problem basically is, "given an input turning machine A and input I, decide whether or not the machine, if run, would ever halt (finish)." It's generally impossible (other than a few simple examples) to figure out without running or emulating machine A. That problem is unsolvable.