Apple ditching PowerPC, switching to Intel

I'm not a big Intel fan but if it would make the slower macs faster, count me in!

There's not much evidence there to support a pretty radical claim.

I think Intel making PowerPC chips is more likely than Apple switching to x86.

I'm going to have to agree with Tom on this one. It seems very unlikely to me that Apple would ditch the PPC at this stage of the game. Moving to x86 would be, in my opinion, a huge mistake at this juncture. If it was ever going to happen, it would have been far more likely when Apple moved from OS 9 to OS X. It just makes more sense, if there is any truth to this rumor, that Intel would start producing PPC chips. I guess we'll just have until Monday to know for sure...

I hope and pray that you are right. I know that Apple has jumped processors in the past, but this one could/would I believe ring the death knell for Apple. They would have to oversee a complete overhaul of all of their own programs--even though Steve Jobs has said that OS X could port to an x86, not to mention the nightmare of all of the developers and their respectives apps. And the article says that he will announce this in his keynote...at the WWDC? He will be lucky to get out of the auditorium in one piece. Even with a couple of years lead time, that is a huge task.

Bottom line though: I swore I would never own another computer with an "Intel Inside" sticker on it again. Do you think that IBM would fulfill individual orders??

Hey. it's no different than when Apple added IBM to the mix and moved from Motorola chips. And if the Chips are better, which I'm sure they will be, I'm all for it.

Hopefully it's not a decision based entirely on financial reasons, but a combination technology amd costs, but more about technology.

Chip technology changes so fast it's hard to keep up and nobody, even Apple can afford to stay with one manufacturer for too long.

The article's best source is "sources familiar with the situation"? I guess it's better than relying on sources unfamiliar with the situation , but still just the same rumour dressed up a bit.

I woudn't go so far as to call that official. Not until Apple or Intel says yea or nay.

It's edge of the seat stuff, though.

I'll wait until I here from one of our engineers, who will be at WWDC..

I will say this, though:

"I'd buy that for a dollar!"

Apple needs somebody to fill in the void left when they couldn't get chips from Motorola. So, they go to one of (the?) largest chip maker and bring them in on the deal. Quite plausible. Most of what Intel needs to do to is just switch some of the manu. and test facilities around a bit and start up a new line. The P4 is dead past 4Ghz as the rumors say, so if they can add another pricey chip to their lines, they'd jump on it.

Apple would be idiotic to switch to x86. Then that means the tight control of hardware goes out the window and commodity PCs will run bootlegged copies of X. Not hardware cash flow and no OS cash flow from those. At least now the can get both, and w/ Itel as a PPC manuf. they will still get both. Finacially, x86 would be a deathstroke.

EDIT: RTA: It says it is to be a "phased transition" from the bottom of the line up. That tells me that is just might be Intel-made PPCs as Tom suggested. They'd have to start the lines at the lower end of the performance capabilites so that the can tune the dies testing facilites to handle better chips. x86 wouldn't require a two-year shift. They can slap together an x86 mobo and get to shipped in under 6 months. THe time frame is all wrong for a simple swithc, and it seems all guesses for going to x86 are guesses in the article. It's just the timeione that throws it off whack, and the financial hurdle of an x86 OS X that compunds it.

Just like Apple's release of the new Newton is imminent and Apple is going to introduce a video iPod in Q3 2003; yeah right. CNet is notorious for causing needless rumormongling.

Cheers,

The Czar

Intel can license the required patents if it has to, but there is no point in Intel fabbing PPC chips. IBM's fabs are just as good, if not better. The reason for the delays is the sheer volume of the chips it has to ramp up to produce for the PS3, Xbox 360, and Revolution. For a product that is used to having 1/10th the amount of sales, I think IBM is doing an acceptable job at juggling fabrication and avalibility.

I think Apple, as a member of the PPC Alliance, could provide the necessary licenses.

I'll believe it when I see it. I'm very skeptical.

It would have made sense 3 or 4 years ago, but not now.

Of all the rumors floated, I think the most likely is that it involves chips for an as-yet-secret device (something like a handheld) or other chipsets.

Everyone get the x86 idea out of your head. It'd never happen.

Intel, if anything, will produce chips for Apple, under Apple's instruction.

Either PowerPC CPUs or chipsets, or quite possibly Intel have invented a very effective graphics card for once(I doubt it) that Apple wants in low-end systems.

But the reason Macs are more powerful than PCs(as a gross generalisation) is the PowerPC architecture. Not the fat bugger of an OS, nor the pretty cases. Apple would be committing suicide if they changed everything(not possible) over to x86 or another sluggish PPC competitor.

And hey- where's Apple or Intel's confirmation that any of this is true?

Also I can't imagine Intel building PPC chips, but I suppose stranger things have happened . . .

dan k

I don't know if this constitutes rumor-mongering or not, so if it does, please forgive me.

Is it possible that Apple wants to incorporate some Centrino-based technology into their mobile gear? I've heard that Centrino is actually pretty sweet in the x86 world. Better battery life and extended wireless range all sound good to me ya know ;-). I can only imagine what could happen if Centrino is paired with the PPC.

As a sidenote, I must agree that it would be suicide to port Mac OS X in it's entirety to the x86. Once everything is transferred over, what's stopping people from running OS X on a Dell? I'm sure there would be restrictions in place to prevent that, but let's face the facts: once OS X is ported to the x86, you can't stop people from running it on compatible hardware. It's like putting up a screen door at a prison. It just won't work.

Here's another thought: The Pentium line is dead after 4Ghz, if the rumours are to be believed. From what I understand, Intel has been having problems with their own RISC designs, and perhaps by producing PowerPC chips for Apple, it will be able to produce them for itself. It is possible the x86 world could be migrating towards PowerPC. We know Windows *can* run on a PPC processor (Windows NT 4.0, anyone?), so it's not completely unfeasible...

Just a thought...

Cheers,

The Czar

I think it's the exact opposite -- it would give Apple the boost it wants so badly. Apple doesn't make much money off of hardware because of such high cost for the components. Software, on the other hand, it can make a killing off of. Once the code is written, the only ongoing cost to sell software is the manufacturing of the media, which is pennies...and with a lot of software being delivered through the Web, cost drops even further.

Apple would be doing itself a favor if it got out of the hardware market. Apple's biggest problem with expanding its user base is that in order to run the Mac OS, you have to buy Mac hardware. Since Apple wants "switchers" so bad, it would be stupid to not port X for x86 machines. In order to "switch" now, you have to drop about $1000 for new hardware -- if Apple ported X to x86, one could "switch" for $100. I guarantee you, more people would be willing to try the switch if it cost $100 than if it cost $1000.

Do you have a cite? I was under the impression they made most of their money off of hardware sales.

Apple has lucritive margins on hardware that other computer makers can only dream of. They would be crazy to give up their hardware business now that they have consumers' interest.

I'm going to make a shot in the dark here but, Intel could have found a way to make the ppc go over 4ghz. It might be way out there but mabey?

What you may be thinking of is that reseller margins on Apple hardware are terrible. Apple themselves makes a decent amount from what I understand.

Centrino is only marketing buzz. It is a Pentium M processor, Intel chipset, and Intel wireless card. Without switching to x86, Apple wouldn't be able to integrate Centrino into their line.

Here's an excellent article by John Gruber on the topic. Don't miss the part about who Christopher Ong is.

The article seems to run down just about all the arguments we've made here. Tommorrow will be the telling, but it'll be a dry day in Redmond if it's true. Anyone remeber SGI? They're running two processor lines, but their OS is also two products. If you want IRIX, you gotta buy the MIPS systems. The SGI Itanium 2 systems run Linux. Even when they made the il-fated x86 NT workstation attempt, they used seperate OSs. SUn is about the only major vendor that I can think of offhand that has their main OS running on two CPU platforms.

Many of the previous arguments make alot of sense, but given the announcement is imminent and speculation is most fun when you know answers will be forthcoming, here are a few speculative rationales for this potentially spectacular but viable move:

(1) Intel (or AMD) refabs the PowerPC for Apple with the blessing and cooperation of IBM because IBM is moving aggressively into services (or in the long - or short - term, maybe Intel (or AMD) actually buys the PowerPC line from IBM...).

(2) Apple offers a proprietary x86-based family of machines and essentially says to IBM and Intel (and AMD): "may the best CPU win"; legacy Apple customers will have longterm (5 year and beyond) support for the PowerPC as it transitions into the Cell architecture, and new Apple hardware customers can start on the x86 with the Apple suite of applications as a base and legacy x86 applications able to run at native speeds (perhaps except graphics) out-of-the-box...

(3) Apple initially attacks specific form factors with x86 based machines specifically designed for them: the server market would make sense, even if just as an x86-based Xserve tossed up against its PowerPC-based predecessor, because networking protocols are (or at least should be) hardware insensitive and most applications brought to servers are either standards-based or user-supported from source (meaning either inhouse development or OSS...). The tablet computer / PDA market is more intriguing and difficult to imagine for such a break with MacOS X based on the PowerPC... but then is that not exactly what they did with the iPod?

I could go on but add just one further idea: the old-time Unix user in me forces me to believe that a well maintained MacOS X has been on x86-based reference machines in Cupertino almost since day-one. The Altivec part was probably tricky and an impediment to any serious consideration, but I understand the IBM-Apple contracts, Apple has 'some' access to that IP along with the basic PowerPC IP. All this is moot, however, if option (1) is the key...

By the way, I finally did buy an iBook (my first Apple purchase) that came with Tiger... and screen rotation works great! Personally, I think a rupture with the PowerPC would be a mistake but that a careful introduction of the x86 could be just the stimulus that Apple would need to grow the installed base...

We will all soon see enough...

Edward

this seems to contradict that rumor:
http://www.eweek.com/article2/0,1759,1823282,00.asp

But as far as Apple loosing its proprietary grasp on HW, no reason they couldn't use x86 chips, and still be proprietary... just because 2 machines use x86, doesn't mean they are completely compatible... After all, Apple didn't go to Dell to get HW. My point is its not just the processor that makes Apple's HW proprietary... its the whole ball of wax.

My concern is losing good developers en mass, like what happened after the Copland fiasco.

But why continue to develop dual core G5's? Who else wants to use them?

If both machines are running x86 processors, then its going to be alot simpler to get Mac OS X to run on a PC. Just look at Wine. Apple may try to prevent it, but it would only be a matter of time before someone hacked a compatibility layer into a PC bootloader of some kind to load the Mac OS on a PC.

Everything you say is true... and the serious move to Unix that MacOS X represents - as compared to the complete reinventing of the wheel that Copland, Rhapsody, NeXT, BeOS, and other efforts represented - makes both worlds possible. Unix offers a hardware neutral environment (with the exception of device drivers... and nowadays that really only applies to graphics... and frankly OpenGL almost removes that barrier). Innovation in the operating system world is as difficult, if not more so, than in the automotive sector: if you don't have something really spectacular to offer that will change the rules of the game, incremental and/or cosmetic change are/is the only way to go...

Anyone reporting from the battlefield yet?

join us!
irc.macrumorslive.com
#macrumors

Just to clear something up. BeOS wasn't an Apple project nor was NeXTstep. Rhapsody was based off of NeXTstep and became the basis of Mac OS X. So, it would be unfair to say that building NeXTstep and Rhapsody were "reinventing the wheel."

http://forum.macosx.nl/album_pic.php?pic_id=7142

The above link is a live feed of the event. Join us in #applefritter on freenode if you'd like to discuss.

Will Apple switch to Intel? Will Steve Jobs make it out of the keynote alive? The suspense!

... on The Register this morning. http://www.theregister.co.uk/2005/06/06/apple_intel_analysis/

They liken it to the move to the PowerPC; if it happens, it'll shake things up for a while, but things will quickly settle down and everybody will be happy. The last line is great: "Five years from now, everyone's going to wonder what all the fuss was about."

Will Tom Owad manage to stay on IRC for 45 seconds without dropping his connection?

Yes, I know that NeXT and Be were only peripherally related to Apple (via Jobs at the first and Gassee at the latter) and should have been more explicit... The idea was simply that a wholly new generic OS will be hard to push without a radical change in form factor or some other overall architecture evolution... OSes for focused apps like gaming and telephony are relatively simple and as such we will soon see whether MSoft has bit off more than it can chew with both the next generations of the mobile device and XBox platforms...

How do I join the applefritter channel? I connected to freenode with Mirc...but it says:

#applefritter You need to be identified to join that channel

How do I get "Identified"?

Thanks

So Apple isn't going to be using x86 architecture, huh?

http://freenode.net/faq.shtml#registering

you'll have to register as we are having problems with the channel

For those who can't get on any of the bogged-down Mac sites:
From Macworld.com:

Since we are outside the priviledged zone (which is only fair for those who paid the price of entry), here is a comment

Windows vs. MacOS X vs. OSS (as in various Linux(es), *BSD, and Solaris) makes for a software-oriented/hardware-agnostic OS war that could give Apple the opportunity to be to Microsoft what Ford was to General Motors in the 60ś: an industrial return to significant marketshare after an illustrious beginning and a rather disappointing growth phase (remember the Edsel?). The dominance that GM had through the 50's evaporated when they (it) failed to adapt rapidly enough to changing needs... and between the Mustang and the first small cars from Japan, their predominance shifted definitively (though it was relatively shortlived for Ford, as we now know...).

That's all I was saying... now back to the news...

Thanks... keep it coming because I have found no other outlet into which to plug... meaning MacWorld is saturated now, too...

Here is the MacRumors.com Live feed stream (I'll gladly delete it if there is an issue, but I wanted a copy perserved if they delete it over there):

My company is now going to have to scramble to see if code that doesn't even work in Tiger yet (due to admitted bugs in the OS) is going to still not work in Tiger for Intel...

grrr...

Steve Jobs, a pox upon you...

It's certainly a cold day in hell! :-)

As I said in the irc channel: Things are different now. I just have a feeling they're going to stay exactly the same.

Also, we now have a decent timeline for when Apple is going to drop support of the Classic environment.

One thing not mentioned: PC architecture. They keep saying Intel chips, but not the PC arch. So, I'm guessing (and also just read a blurb on /. ) that they might have/planned an x86 mobo version that uses OF and not a PC BIOS. That would be one step to prevent quick piracy, but if it runs on x86, then some mods to VMWare and they can get it running.

If it really is the regular PC arch. then it IS a cold day in Redmond. Good by Winders for anything I've got!

It looks like they are making plans that go further than I had hoped. I still do not see why they must drop the PPC while adding x86... and is their dependence on Intel any different that the previous dependencies on Motorola and, subsequently, IBM? And there was apparently no mention of AMD (though that was probably out of deference to Otellini...).

And I am not a fan of "automatic binary translation": anybody out there remember ANF (Architecture Neutral Format) or follow the Itanium saga (and the latter was Intel's baby, after all)?

I guess I just don't understand how these processor deals are made because it would seem like they could have "simply" said "IBM and Intel/AMD, you've got three years: may the best CPU win...". Apple clearly wants to continue to make its own machines, so the 3rd party/whitebox option is not there... so why drop the PowerPC so precipitously?

I guess we'll see... Overall, I think this creates some problems for Microsoft (I doubt they will port Longhorn to Cell and drop Intel...) and perhaps enhances the opportunity for Linux and its OSS brethren...

...wouldn't it be nice to wake up in three to five years and basically have a pie cut into three equal pieces, each with advantages and disadvantages that would translate as choice?

Quick thought: DOes this mean that the mini was in design so that they could get the last stocks of G4 chips in the wild before they announced the switch? That Apple may have had some legal wranglings to buy a certain amount of chips for their systems, and the mini design was a good way to meet quotas before the hardware took a sales hit from the annopuncement? Dr. Bob stated that the mini was on the drawing boards when he left Apple, and the x86 conversion has been worked on for the last five years. So, if Apple has a good roadmap for the transition, thent he mini might have been the perfect device to buy out of their old G4 contracts, and also have a good form factor to make the x86 an otherwise desirable switch, a-la the AOpen/Intel clone.

"After Jobs' presentation, Apple Senior Vice President Phil Schiller addressed the issue of running Windows on Macs, saying there are no plans to sell or support Windows on an Intel-based Mac. "That doesn't preclude someone from running it on a Mac. They probably will," he said. "We won't do anything to preclude that."
However, Schiller said the company does not plan to let people run Mac OS X on other computer makers' hardware. "We will not allow running Mac OS X on anything other than an Apple Mac.""