What is the Intel Switch Costing Apple?

SenseOfHumor writes "A Business Week article says that it costs Apple $898 for an Intel iMac before loading it with software and packaging. From the article: 'But for Apple, the switch to Intel chips is less about saving money in the short term, and more about hitching its wagon to Intel's longer-term product road maps, particularly in the area of notebooks. IBM's chips are power-hungry and generate a lot of heat, and therefore not suitable to notebook computers.'"

Comments

If they don't know....

[MountainMan101]

If they don't know, why ask us? Everyone knows slashdot crowd knows nothing. But we'll always comment. So I'll say it's costing them at least a hundred pigs a month in tribute. Maybe some biscuits (you Yanks call them cookies).

Re:If they don't know....

[Zerbs]

Yes, but how many Bacon Jr. Cheeseburgers is that?

When did this change?

[Anonymous Coward]

In the mid 1990s, Apple showed the famous picture of a Pentium grilling a hot dog and claimed Intel's chips were power hungry and ran hot compared to the nice cool sleek PowerPC. That was one of the supporting reasons that Apple ostensibly switched, according to all the engineering presentations at WWDC. So when did this change?

The main reason of course was that RISC processors were on a much faster performance incline than the fuddy duddy old CISC processors like the x86 line. The graph comparing the two in the period 1995-2005 showed CISC acceleration continuing to slow and RISC acceleration continuing with, I believe, a skyrocket attached to the top of the graph. We all know how that turned out.

Re:When did this change?

[adisakp]

The main reason of course was that RISC processors were on a much faster performance incline than the fuddy duddy old CISC processors like the x86 line. The graph comparing the two in the period 1995-2005 showed CISC acceleration continuing to slow and RISC acceleration continuing with, I believe, a skyrocket attached to the top of the graph. We all know how that turned out.

No one at the time expected the changes in CISC processors. CISC processors still do have a "complex" instruction set in that they allow multiple forms of adddressing and varying length opcodes. However, internally these chips have become much more RISC-like. The current generation of Pentiums actually does an internal version of dynamic translation from CISC to RISC-micro-ops (which may be 1 or more per CISC instruction) and executes the micro-ops using a different instruction set internally. This internal RISC instruction set is used so central to the design that the L1 I-Cache is not actually a verbatim data cache of the CISC instructions but actually a trace cache of the translated RISC-like micro-ops.

Re:When did this change?

[dasil003]

No one at the time expected the changes in CISC processors. CISC processors still do have a "complex" instruction set in that they allow multiple forms of adddressing and varying length opcodes. However, internally these chips have become much more RISC-like. The current generation of Pentiums actually does an internal version of dynamic translation from CISC to RISC-micro-ops (which may be 1 or more per CISC instruction) and executes the micro-ops using a different instruction set internally. This internal RISC instruction set is used so central to the design that the L1 I-Cache is not actually a verbatim data cache of the CISC instructions but actually a trace cache of the translated RISC-like micro-ops.

It really just goes to show the error in the view that RISC and CISC are considered opposite approaches to processor design. The dichotomy was more pronounced in the early days of chip design, but the fact was that proponents of both approaches had good points, and so it was inevitable that modern chips combine the best of both philosophies.

I think the progress made on the PowerPC architecture is a testament to its viability. The fact that it's even managed to stay anywhere close to Intel/AMD is remarkable given the difference in R&D dollars (I'm just guessing). But the timing of the Intel switch makes perfect sense.

Consider the switch to the PowerPC in the 90s. It was a time when Microsoft was rapidly catching up to the Mac in terms of UI, and computers were generally underpowered for the common applications that people needed. Gambling on a more promising architecture could have paid off huge if the performance panned out. That never happened, and Apple was in pretty bad shape by the late 90s.

Now, however, computer performance has reached adequate levels for all the things the common people want... audio, video, web surfing, word processing. We can always use more power, but performance is not such a big deal as it used to be. Since they're not seeking a competitive advantage in performance, it makes sense of Apple to at least assure commodity performance by going with the dominant CPU architecture. Apple has contiunously struggled with supply problems from chip vendors for years, hopefully this will now be behind them, and they can focus on the creative part of their business which is where they've always excelled.

Re:When did this change?

[Overly Critical Guy]

Since they're not seeking a competitive advantage in performance, it makes sense of Apple to at least assure commodity performance by going with the dominant CPU architecture.

The Intel switch wasn't about switching to a dominant architecture, it was about moving to a platform that had a future roadmap for performance-per-watt. Intel is kicking butt in that department with the Core Duo (a laptop chip that manages to compete with a desktop Athlon64). Merom and Conroe later this year are supposed to further this even more dramatically, being chip redesigns with performance-per-watt as the design goal.

Steve Jobs was tired of selling a G4 Powerbook, so he moved to Intel.

Don't We Know this already?

[patman600]

Wasn't this the publicly stated reason for switching when Steve announced the move last summer? They said IBM makes great server chips, but the future of personal computing is laptops, something Intel is putting more R&D into than IBM, and thus provides a better solution.

why is this news?

Re:Don't We Know this already?

[AugstWest]

The news is about the cost per iMac, but this being /., everyone is focusing on the reason for switching, since it has already been rehashed a thousand times and they're comfortable flaming about it.

Really, what this article is saying is that Apple is only making $450 per low-end iMac sold, based on their own estimates, which are most likely wrong.

Why is THAT news? You got me.

What about the possibility of avoiding lock-in?

[Yhippa]

It should be easier to switch to AMD or other X86 platforms in the future, opening up more negotiation possibilities.

Power grab

[digitaldc]

"IBM's chips are power-hungry and generate a lot of heat..."

In a related news item, IBM chips are now running for elected office worldwide.

All Intel, All The Time?

[SoupIsGood Food]

This is why I'm confused about the push to "All Intel, All the Time!" Apple, with Mac OS X's Unix and NeXT roots, should embrace a multi-platform strategy to get the most bang for its buck wherever it can. The PowerPC-derived Cell will rock for workstation and servers, and the Meron will kick major butt for home user kit. Best tool for the job, and just compile for the famous NeXT "Fat Binary." Back in the day, the same NeXT executable would run on 68040, Sparc, PA-RISC and Pentiums. Why not now? Why tie yourself to x86 alone, when there are better alternatives to fit the niche you're targeting?

Too much politics, and not enough engineering.

~ SoupIsGood Food

Re:All Intel, All The Time?

[Zathrus]

The PowerPC-derived Cell will rock for workstation and servers

And this conjecture is based on what? Certainly not any real world evidence. The Cell architecture is completely untested and a radically new design for a commodity chip. On paper it looks decent, but so did Itanium (and, technically, Itanium is quite good... except that the software has never been able to properly exploit it. Much the same may be true for Cell).

Back in the day, the same NeXT executable would run on 68040, Sparc, PA-RISC and Pentiums.

Yeah, and look at what a fantastic success story that was! I mean, we have NeXT cubes everywhere now!

Frankly, it's a drawback. Software developers have to certify their code on multiple platforms if you do that, and that's hideously expensive. Sure, you can claim that you can compile for one and it'll work on them all, but know what? That's a lie. It's always been a lie and it always will be one.

Writing cross-platform code isn't as hard now as it used to be, but it's still not trivial. Even if you're talking about different CPU architectures on the same OS. We see that at my workplace running on HP-UX when it comes to PA-RISC vs Itanium; we develop on PA-RISC, but some of our customers run on newer hardware and we cannot replicate the bugs. And this is for software that compiles on several flavors of Unix and Windows, across more CPU types than I care to list, and under at least 3 different compilers. We've already done the hard work of writing cross-platform, cross-OS code, and yet same-platform/different-CPU bugs still happen.

You throw that kind of crap at your average development house and they'll do one of two things -- only develop for the most popular configuration (thus helping to marginalize the others) or just develop for another platform that doesn't have these issues (e.g. -- Windows and x86).

Economy of the 'Change'

[palad1]

I don't know what it costs Apple, but I sure know the change to Intel will cost me about 2000 .

Re:It's the marketing angle perhaps?

[Ryan Amos]

Everyone forgets, but Apple has had a smooth transition between architectures before. They moved from the Motorola M680X0 architecture to the PowerPC by using mixed binaries, and had very few problems. There were some initial growing pains (extensions that would bomb the system, etc,) but by and large the transition went smoothly.

And that was on System 7; OS X is a much more portable operating system. A simple recompile is all that's necessary for most programs without a lot of assembler optimization.

They'll maintain differentiation with case design. Don't expect Apple to ship ATX systems; they moved to Intel because laptops are quickly becoming the standard, not desktops. Every laptop manufacturer uses custom designs anyway, and the IBM chips were really designed for servers and workstations (the POWER line at least,) not laptops.

One bonus is that they no longer have to emulate the x86 to do windows emulation, just translate the APIs. Apple has also written stuff like this before; with Classic mode on OS X. In 2 or 3 years I wouldn't be surprised to see Windows .exes run under OS X as if they were native applications.

Apple has their foot in the door of consumers' wallets/minds with the iPod. Now that everyone and their mother (literally) has an iPod, they'll be more open to purchasing a Mac as their next computer. With users becoming increasingly fed up with viruses and spyware, Macs are a very attractive option to many people. Once the price comes down a little bit (which I suspect it will once they ramp up full scale production on Intel) I see nothing but good things for Apple.

Re:the real costs

[cbackas]

It might be worth noting that Intel based Macs will *not* support Classic mode in any way. I seem to recall reading a knowledgeable article saying this directly, but I can't find it right now. However, if you refer to http://developer.apple.com/documentation/MacOSX/Co nceptual/universal_binary/index.html#//apple_ref/d oc/uid/TP40002217 you'll see that both older applications AND Classic itself are listed as things that Rosetta can not run. If Rosetta cannot run a Classic app, then it's not GOING to run as it's still PPC.

Now, I'm sure emulators will eventually appear, but this isn't the best example to present to demonstrate Apple backward compatibility =)

Re:the real costs

[Jerry Smith]

porting operating system $30,000,000

OS X derives from NextStep/OpenStep, and has been developed for n86 from day 1. They ported every release to PPC. Yes off course it needed aftercare, but still: first for n86, then to PPC.

Re:Uhmmmm

[SysKoll]

The G5 was a decent chip, IBM just didn't have a mobile chip to sell Apple and was too distracted by Xbox 2 and PS3 to care.

Very true. Volume-wise, the game console market beats Apple's meager volume hands down. No wonder then that IBM chose to devote its Microelectronics division's resources to making the PowerPC derivatives for the Nintendo Revolution, XBox 360 and PS/3. Not to mention embedded versions you find in consumer items and under the hood of cars. The Cell processor alone will find its way in many consumer electronics appliances, not just the PS3.

So the choice was between making a laptop chipset for Apple (volume: hundreds of thousands a year) and making a high-volume chipset for several consumer markets (volume: millions a year). Guess where IBM prefered to invest. Can't blame them for telling Apple to go fly a kite.

I completely disagree

[Critical_]

Just within the last 12 months has Intel started releasing chips that focus on lower heat and power.

False. Your statement isn't giving Intel enough credit and is not supported by the numbers. Since the original Banias Pentium M's were released back in March of 2003, we've seen Intel's mobile products have very good performance per watt ratios and overall power usage numbers. In fact, the overall power usage was the lowest in the original Pentium M's out of the entire line. You statement would be correct if you it said this: "...within the last 34 months (i.e. ~3 years) has Intel started releasing chips that focus on lower heat and power."

Data pulled from Intel Product Specifications at http://www.intel.com/

Banias (the normal voltage models-i.e. 1.7 GHz, 1.6 GHz, 1.4 GHz, etc):

Thermal Design Power: 24.5 W (Full speed) / 6 W (Speedstep)
Sleep Power: 1.7 W
Deep Sleep Power: 1.1 W
Deeper Sleep Power: 0.55 W

Dothan (any model #):

Thermal Design Power: 21 W (Full speed) / 7.5 W (Speedstep)
Sleep Power: 3.2 W
Deep Sleep Power: 2.5 W
Deeper Sleep Power: 0.8 W

Core Duo (any standard power model #):

Thermal Design Power: 31 W (Full speed) / 13.1 W (Speedstep)
Sleep Power: 4.7 W
Deep Sleep Power: 3.4 W
Deeper Sleep Power: 2.2 W

The Pentium M chips were a step towards lower power, but the Intel Core Duo that ships in the imac is the first chip that is really ahead of AMD for mobile systems.

Again, False. The first part of that sentence has already been proven false with the numbers I've posted. The second part of your AMD fanboy'ism is also incorrect. AMD offers two TDP ranges in their "Lancaster" single core Turion64 mobile processors: 25 watts and 35watts. As you can see with the data presented above, both of these TDP's are larger than Intel's single core Pentium M offerings which have been available since March 2003. AMD's Turion didn't even arrive on the scene until 2005 which gives Intel a solid two year headstart. What's even more interesting is that more than half of AMD's entire single core Turion line consumes more power than Intel's dual core Core Duo mobile processors. AMD has yet to release their dual core Turion processors. So your statement that the Intel Core Duo is the "first chip that is really ahead of AMD for mobile systems" is complete wrong. Intel has had AMD beat since March of 2003 in the mobile market and still continues to beat it. Please check your facts before posting lies or put an AMD fanboy disclaimer on your posts.

Note: I didn't both including Intel's various Low Voltage and Ultra Low Voltage Pentium M, Core Solo and Core Duo processors that have an even lower TDP than the standard voltage processor numbers I posted above. Adding this information would only serve to futher prove that your statements are wrong.

Re:The finite choices come from infinite options

[adrianmonk]

Apple is not a software company. They are a hardware company. It's that simple. They build really solid, nifty hardware that apparently reaches fetish level for a certain market, and they've learned to turn that market into money.

I would argue that they're not a software company, but they're not a hardware company either. Instead, they're an integrated system company. Years ago, before the PC and Windows (and Linux, which has the same model) took over, you bought both an operating system and a computer. The two were pretty much inseperable. (This was how the IBM PC started out, as well as the Mac, the Amiga, the Atari ST, the Commodore 64, the Apple ][, etc. And the same thing was true before personal computers: VAX machines had VMS, IBM machines had one of IBM's 99 different operating systems, etc.)

These days, not as many people are doing the same thing. Certainly if you buy a machine from Dell, Dell is working with Microsoft to make sure the system has all the right drivers. But that's not quite the same thing as an integrated platform where hardware design and software design are done by the same organization. Integrated hardware and software designs are available from Apple and also a few other companies like Sun. And the interesting thing is that both Apple and Sun have now adopted some x86 chips. Sun has Opteron servers and workstations available but continues to make new SPARC chips (including Niagara, a whole new series of chips), and Apple is using Intel chips in desktops and laptops.

For what it's worth, there is some value in an integrated system. Knowing that all the hardware and software come from the same place gives you a greater degree of confidence that it will all just work together. And if it doesn't, when you call for support, you are dealing with only one organization, so the blame game ("it must be the other vendor's product, not ours") is less likely. A certain percentage of the people are willing to pay a bit of a premium for these advantages, so that gives Apple (and Sun) a market that is a bit different from the regular market, which gives them a niche to play in.

Of course, it doesn't hurt that Apple has really snazzy industrial design and that people look at an Apple laptop and instantly want one without yet even knowing what's inside. Think of the amount of appeal PowerBooks have had for the last few years even despite the fact that they still contain slow G4 processors.

Why Intel? Jobs wanted Intel. Why now? OS 9.

[argent]

Apple has been trying to kill the classic Mac OS and replace it with NextStep, I mean OpenStep, errr, Rhapsody... since 1997. The original plan was for all new development to be in what's now Cocoa and was at the time called Yellow Box, and legacy apps would run in a simpler version of Classic that basically ran a whole OS 7 or 8 session in a single window, called Blue Box.

The ISVs, paricularly Adobe, plotzed. There was a major row with threats of abandoning the platform, and Apple backed off, improved Classic, came up with Carbon as a transition API, and brought out OS 9 and eventually OS X.

Steve Jobs reportedly had wanted to go with Intel as soon as possible. He thought Apple had made a mistake switching to the Power PC while he was away at NeXT. OpenStep ran on Intel, of course, and Apple had versions of Rhapsody that ran on Intel boxes, even on generic clones. They had a fat binary mechanism in OpenStep that supported by the end as many as five different processor architectures.

And that's why intel. Not because IBM screwed up, but because it was in their long term roadmap and had been for years.

But obviously... that wouldn't fly if they couldn't even cram classic Mac OS off in Blue Box.

But they kept their Intel code base alive, and every other year, about, they tested the waters by trying to stop offering a Mac that could boot up into OS 9.

Every time there was a user revolt.

Until late 2004. The last G4 that could boot to OS 9 disappeared from the Apple store, without any fanfare. And, apparently, there just weren't that many people dependent on OS 9 to make enough noise to notice.

A little over 6 months later, they announced the Intel switch.

Rosetta will run all legacy Power PC applications... well, all legacy Carbon and Cocoa applications that run on OS X. They're not running Classic under Rosetta. Classic is dead.

And nobody's bitching about that, either. Which means they guessed right, and Apple can finally drive a stake into the heart of Classic Mac OS and leave it behind for good.

And that's why they did it now. Because they could.