Slashdot Mirror
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.'"
It should be easier to switch to AMD or other X86 platforms in the future, opening up more negotiation possibilities.
The article didn't mention overhead. You can bet that there is a cost associated with the overall organization, plus the physical plant, R&D, etc. that most likely brings the costs way up from where the article puts them!
"Money is truthful. If a man speaks of his honor, make him pay cash." Notebooks of Lazarus Long, Robert A. Heinlein
...Intel gets dumped in favor of AMD?
GetOuttaMySpace - The Anti-Social Network
It is odd to me that Apple leverages so much into specific processors rather than specific processes. It would seem to me that Apple really has a great interface -- and that is the product they want to sell. With their OS kernel being based on some *nix variety (BSD? I can't remember) I would guess that the processor itself is unimportant if their software and APIs are hardware transparent.
Here's the great thing about the market and letting it lead you (instead of the other way around) when you are an OS or software provider -- you can focus on writing good clean code, and follow up that code with the hardware that offers your code the absolute best package given the infinite choices.
Power management, heat creation, MIPS, FLOPS, BOPS, GHZ, THZ, MB, MBps, whatever the hardware does best, there's always a ratio to price. That's the great thing about the free market, though, competititors will always want to beat the other.
What is stopping Apple or another software company from offering the best darn interface for programmers and users to work with, and then find the processor to wrap the interface around? Is this Apple goal with Intel, possibly? Shake up IBM (and show smaller processor companies that they, too, have a chance) and create an operating system that must now work with 2 (or 10?) completely different processor subsystems? Is this Apple showing that they can get away from hardware entirely, and focus just on software?
The pentium-m processors are incredibly power efficient and perform very well. Sure there desktops are absolutely horrible from the Northwood to the Prescott core (and perhaps some new cores since i've stopped paying attention to what intel releases on the desktop now) but that doesn't exclude the fact that they do infact have one of the best, if not THE best solution for notebooks.
Hmmm... Pie...
This wasn't a knee-jerk reaction, Appple was unable to build a fast laptop, and IBM couldn't offer them anything competitive with what was happening on the x86 side of things. I've got the latest Powerbook G4, which is the best, fastest laptop Apple could offer until now, and it's just too far behind the curve. Would you rather they remained there, while IBM worked on other things and didn't care?
I'm a Mac user, and I've been keeping an ear to the ground, but I haven't heard any mention of the new MacBooks having improved battery life over the 'old' PowerBooks, so I am guessing the reverse is true (or much would be made of the better battery life). Of course, there are lots of other reasons for the move than just lower power consumption, and even on that front, there's no way of knowing right now if the new MacBooks will have lower unit-of-power/unit-of-computational-power costs. With the possibility that the new chips provide better-than-G5 performance in a laptop, well, there's certainly something going right with this switch, even if Intel doesn't have the best reputation for efficient, cool chips.
Although the moon is smaller than the earth, it is farther away.
So when did this change?
Somewhere in the last decade where each architecture was developed into something different than it was.
If Apple's going to be commodity CPU on the server front, then there's no incentive on the hardware front to pay for Apple.
Uh, why do you say that? You're saying that the only important hardware consideration for a server is what brand of CPU it uses. All Intel servers are otherwise equally desirable, and all AMD servers are otherwise equally desirable.
That's obviously not the case.
And really, no one in the past five years bought an Apple because of the PowerPC processor. They bought one despite it, because the hardware was great otherwise, and because the OS was great.
This space intentionally left blank.
The Core Duo is a great laptop chip, have you seen the benchmarks and reviews lately? IBM had no real roadmap for a laptop version of the G5. Shortly after the switch was announced, IBM made some vague statement saying that they had a low power G5 design, and they could have made it if Apple wanted it. I seriously doubt that their chip would have come near the performance of the Core Duo, or that it would be ready today.
The CPU benchmark numbers tell the tale. The Core Duo is 4-5x faster than the 1.67GHz G4 in the PowerBook, but only 2x faster than the single-core 1.8GHz G5 in the old iMac. So you can assume that the Core Duo is at least twice as fast core-for-core as the G4, but about the same core-for-core as the G5.
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.
- "When you want something with all your heart, the entire universe conspires to give it to you" -Paulo Coelho
"What is stopping Apple or another software company from offering the best darn interface for programmers and users to work with, and then find the processor to wrap the interface around?"
I think the problem is that Apple is a software company that makes its living as a hardware company. And to make money from hardware, they have to be perceived as different from their competition. If you follow what you're saying to it's logical end, you come up with a solution that says "Apple should not sell hardware, they should write software that runs anywhere".
I'm sure Jobs experience with NeXT tells him that selling an operating system, his experience watching Gasse sell BeOS tells him he doesn't want to compete with Microsoft on that basis. So he's chosen a middle ground that appears to be increasingly difficult to maintain differentiation on the hardware side.
The next few years will be interesting for Apple, that's for sure.
You were mistaken. Which is odd, since memory shouldn't be a problem for you
Why would the general populace ever want to buy a Mac?
Almost every person that I've induced to switch or helped to switch were prompted to do so to escape the Windows virus nightmare.
Build quality , support quality and OS/Software quality
The only things certain in war are Propaganda and Death. You can never be sure which is which though
Apple decided years ago that laptops were going to be the future, and the age of giant towers was coming to a close, and odds are that's true.
Small, lower power chips that put out decent numbers are worth more to most people that large, power hungry chips that put out phenominal numbers. It's funny, the story below talks about AMD chips outselling Intel chips in the desktop. At the end of the day though, I fear AMD is taking over a market segment as it's being abandoned, nothing more.
I'm just waiting for the notebook manufacturers out there to start cloning the Apple machines, and stick a cooler processor in with a bigger battery. The specs for the Apple machines aren't unknown, and they are using mostly market pieces.
Yeah, I won't have an Apple that lights up, but I won't be paying the Apple toll for the same hardware either. And, chances are, I'll be able to use OS X.n anyway.
Linux - because it doesn't leave that Steve Ballmer aftertaste.
Luxury auto brands like Acura, Infinity, Lexus, Cadillac, and Lincoln should all be folded, because they are super dumb. I mean, who's going to pay the "Cadillac Tax" just to get a glorified Chevrolet? Corporations aren't going to switch; our fleet here doesn't even SEND requests for bids to any of these brands--they go to Chevy/GMC or Ford, because they provide us with machines that get the job done. And sure, all the Cadillac fanbois will tell you that the user interface is so much nicer than a Chevy (even if underneath the gloss they are indistinguishable), but at the end of the day who really cares? Price/performance is all we look at.
In conclusion, Apple will definitely go out of business, just like luxury car brands, because nobody in their right mind will pay extra for something nice.
Repetition does not transform a lie into the truth. - FDR
With the understanding that this was not intended to be an accurate estimate, it's still an exaggeration.
Apple has, at the very least, shown it's operating system to be more flexible (or morphable) than Microsoft's Windows.
Apple already made a successful platform jump once before (from 680x0 to PPC) and maintained side-by-side compatibility.
They've also now made an operating system jump (from the MacOS 9.x series to the OSX series fully encapsulating MacOS 9.x as Classic.)
I think the best Windows ever offered was a version of NT which ran on a Sun box. That didn't last too long.
So they've already paid the price for freedom from hardware lockin. Now they're just cashing-in.
I look forward to testing some old System 4.1 apps in Classic under OSX on Intel. From everything else I've seen, support should be transparent.
Which, when you think about it, truly is priceless.
The thing about things we don't know is we often don't know we don't know them.
Did you ever think that it was because there is no currently supported version of WINDOWS for the PowerPC or MIPS or ARM or Sparc... or ANYTHING that is not X86?
Nothing in the world is more dangerous than sincere ignorance and conscientious stupidity.
On thing that is not called out in this article (at least not well) is that Apple is saving R&D costs and R&D time by not having to develop its own chipset like is has done in the past. Instead Apple is utilizing Intel developed and manufactured chipsets. Intel has the economy of large volumes for their chipsets, Apple did not.
When Apple was making its own chipsets they could only afford to revamp them every couple of years because of the low volumes in relation the development cost and manufacturing tooling and ramp. Now Apple can refresh their chipset and product offering as often as Intel does without excess cost.
The component costs per unit may be higher but saving in both time and money other places will help make up for that.
If it becomes trivial to pirate the OS, and a significant number of people do that instead of buying a mac, it costs them their business. If it becomes trivial to dual boot Windows, and people do that, and then developers stop developing for mac and tell them to just boot windows... it costs them their business. It's much more than just a price point.
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).
Apple has, at the very least, shown it's operating system to be more flexible (or morphable) than Microsoft's Windows.
Apple has always kept the Intel jump as an Ace card in its back pocket. Rhapsody was developed for both Intel and PowerPC, and Apple kept Darwin x86 up to date. For many of us, the only surprise was that Apple actually made the jump, not that they could do it.
Javascript + Nintendo DSi = DSiCade
Apple already made a successful platform jump once before (from 680x0 to PPC)
More importantly, NeXTStep made the jump to Intel in the past. So OS X already has a history of running on x86.
It's also to enable 'pro' users currently produced Powerbooks because not all software has been moved to universal binaries, and a lot of pro-stuff won't run fast enough using Rosetta for these users. The PPC machines will be around for a while, just as OS9 machines were available.
Anyone who whines about being modded down should be.
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.
--
Mad science! Robots! Underwear! Cute girls! Full comic online! http://www.girlgeniusonline.com/
The performance for a given power crown has been handed back and forth for a while between Intel and AMD. While it is true Intel has had Pentium M's for quite a while, they have not been comparable to competing AMDs for performance for most of their existence, barring a few anomalies. This is my unbiased opinion. I am neither an AMD not Intel "fanboy" as so many on Slashdot seem to be. I haven't yet purchased a non-PPC laptop in this millennium. Looking at arstechnica or a similar sites comparisons over the last few years seems to show that most review sites agree with my assessment. To summarize, your assessment is completely correct, if you don't care about performance as part of the equation.
Intel has the fab capacity to handle any requests Apple makes. AMD doesn't. If Apple went to AMD, they would instantly become AMD's biggest customer.
That's an interesting assertion, considering that AMD processors are outselling Intel processors in the retail marketplace in both retail box and OEM PC form.
In that respect, it's quite similar to the Itanium (no hardware branch prediction, all in compiler) -- screaming fast for something that's very well optimized, but change the processor (Itanic II), and you get bad performance on code compiled for the first rev of the chip.
Now, for the specific case of Photoshop, the cell might work quite well -- as a coprocessor, for filters and other ops in Photoshop, but not to run the main UI. Same thing for something like Mathematica or Matlab. It's not a very good general-purpose core (in terms of being easy to program or easy to get good performance).
It will show in the PS3 -- the first games will have horrible performance compared to games that come out six months later, as the developers understand the oddities of the Cell better and the tools get better. Yes, that difference shows up in every console that comes out, but I think it's going to be especially pronounced on the PS3 because of the cell.
-30-
I'll make a prediction that part of the Apple/Intel deal was two-fold on the Firewire 800 situation. First, Apple agreed to dump FW800 to play up Intel's USB technology. That's nothing new, but second, Apple will collaborate with Intel on making a USB 3.0 that's at least as good as FW800 if not better.
I doubt that Apple's move to Intel had a great deal to do with performance, and I dislike this fact being used as a key selling point for the iMac. If you refer to the "definitive" G5 vs. everyone else benchmarks at http://www.anandtech.com/mac/showdoc.aspx?i=2436it is apparent that the G5 is largely comparable to offerings from AMD and Intel (admitedly the new Intel Core Duo is not benchmarked) and although the G5 is, in many cases, not the fastest chip, it is similar. The increases of 2-3x in performance between the G5 and MacIntel iMac are a consequence of having a dual core chip (and being a generation ahead of the G5) besides, Apple could have feasibly used the dual-core G5 chips that they've had at their disposal for a while now. Any Mac zealot will argue that their PowerPC Mac is "just" as fast as an intel based system, but performance is NOT the issue. This is why the iMac was updated first, it is a consumer product, supporting Apple's fledgling attempts to enter the living room (consider front row ) - it desperately needs Intel's brand name associated with its hardware.
The significance of this new product is long term and cannot be underestimated.
Apple finanlly has penetrated the consumer electronics market with the iPod, and their brand recognition and image could not be better. Apple has shoehorned its way into the psyche of the common man. It now has to bring its key product, the mac, to the masses. Consumers will be attracted from a design perspective and because it shares the same logo as their iPod, the OS is a little different to windows, but now at least you have the reasurrance of dual booting into windows (I'd like proof of this concept, but I'm sure it will come) and the processor gives the security of a well recognised brand name (consider brand strengh of Intel vs. AMD).
In the future, I doubt that IBM's die shrunk Power chips will share the low power consumption that I expect Intel will bring, and many concepts for great products will never be realised. I'll be interested to see if the new Intel chips can match up to the PowerPC altivec-ised vDSP FFT's , but in a way I don't care. It is an exciting time to be a Mac user, as more people join the fantastic experience that we have had for so long, and new software and hardware comes our way. Either way, they're finally here and it will be interesting to see what the future holds.
Nothing sucks like a Vax, nothing blows like a PowerMac G4
Classic is just OS9 emulation to cover people during the change to OS X... It's time to let it go.
Someday a real rain is gonna come...
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.
"Sufferin' succotash."
That can't be right, AMD sells into HP, IBM, and pretty much everyone but Dell's computers, and we all think Dell will see the light soon. Any of those outsell Apple. I'd bet retail processor sales even outsell Apple. Capacity is easy, particularly as Apple is going to commit to one and only device (for now).
The fact is you won't know why they chose Intel because anyone who does know can't talk about it. I'm not implying that it is necessarily something unethical, just unannounced and lawyerized.
The argument that "performance has reached adequate levels" rears its head every few years or so in the industry. The fact is though, everything goes in cycles. Sometimes the software development outpaces the currently available/reasonably priced hardware, and then things shift back the other direction for a little while. But the one thing that's certain is; development isn't going to come to a halt on the software side. If you develop faster, cheaper systems - eventually, software developers will figure out ways to make use of everything that's available to them. They have to, because in most cases, that's the only thing that keeps food on their tables. New versions are expected practically yearly for most popular applications, and once you've offered all the basics - what else is there to do for the next upgrade? You have to add "cool new things" that catch people's interest. Whether that means toolbars that automatically fade into the background when they're not used for a little while or voice recognition integrated into the app, built-in video tutorials or adding all new capabilities to perform tasks the app never tried to tackle at all before - you're going to need ever faster CPUs to become "commdity items" to go along with your work.
Apple has a deep hole to keep trying to dig themselves back out of largely because the perceived "value for the dollar" of buying a Mac became VERY poor in the mid to late 90's. Sleek new systems running OS X have started turning things back around - but Apple's move to Intel means they've got to be MORE concerned with performance increases than ever before! They can't lean on an excuse (however accurate or inaccurate is really was) of "You can't compare Mhz to Mhz between Intel or AMD chips and our PPC chips!" Now, the CPUs powering their hardware are the SAME ones powering everyone else's hardware. So if your new Mac offers a 2.1Ghz CPU and a new Dell has a 3.0Ghz of the same product type - it's clear. The Dell is a lot more powerful. And the general public understands that.
I recall that in the past, Intel and AMD used different numbers for "Thermal Design Power" and they weren't directly comparable. You are comparing them here. Have they changed -- do the numbers compare now?
AMD always published worst-case TDP numbers: the most heat the chip would ever dissipate. Intel preferred to publish "typical" numbers.
Also, the Intel chips would throttle themselves down to half-speed if they got really hot, so they would never reach their theoretical top temperature; the AMD chips didn't used to do that. (And still don't? Not sure.)
My understanding is that currently Intel is beating AMD for low-power low-heat chips, but AMD has some stuff coming that will be competitive or even better. (Of course, Intel totally creams AMD on ability to supply large quantities, and especially on the ability to guarantee a supply of large quantities, so Apple would have gone with Intel even if AMD had the best low-power chips now.)
What I want is a workstation that runs silky smooth and is totally silent. I really want an ATX motherboard with a notebook processor and a truly giant heatsink with a slow-moving, giant fan. Couple that with an nVidia 6600 board with a giant heatsink and no fan, and a power supply with no fan, and stuf a few GB of RAM in and make it net boot (no hard disk). For more mainstream use, put in a flash drive and make it boot from that. (Not everyone has a server in their house to boot from liek I do.)
And when it comes to software, Apple has no peer. Apple consistently creates great applications that normal people want to use. Apple's competition, on the other hand, has demonstrated -- repeatedly -- that they cannot do the same.
So that's the reason for the switch to Intel. Apple has moved what used to be a two-front war onto a single battlefield where it has the ability to outmaneuver all opponents.
Smart move. Expect Apple to capture some market share.
Easy, automatic testing for Perl.
The incremental cost of putting software onto a system is close to $0. At least for code they wrote themselves. Licensed code will have a small incremental cost.
Note that the article only covered incremental costs, i.e. the amount it costs to make a machine out of raw materials and labor. Software costs are almost entirely based on development costs. Development costs are of course real, but a lot harder for outsiders to estimate. Note that they didn't even attempt it for the hardware side. But development costs can be divided amogst all the many items sold; incremental costs apply to each item sold.
Software sucks. Open Source sucks less.
Assume Apple shipped 50% laptops, the volume would be 2 millions a year. Say 5 millions over the next 2 years (expected life of such a chipset design). Not bad, but we are talking about an engineering effort costing $50 to $200 millions. The resulting chipsets would have been $10-$40 more expensive than Intel's in order to pay for this effort, and that's before production costs kick in. You're starting to see why it wasn't a good deal even for Apple.
Now, after its move to Intel, Apple benefits from a standard PC laptop chipset that will be sold in huge volumes (and for which development costs were footed by Intel, not Apple customers only), and IBM can focus on the consumer electronics market where the volumes are in the tens of millions a year.
Volume, thy law is cruel.
--
Mad science! Robots! Underwear! Cute girls! Full comic online! http://www.girlgeniusonline.com/
I didn't mean to imply they achieved parity with Alpha in their first iteration, but early Pentium Pro and II put Intel on a pretty close footing with MIPS and SGI in particular, especially the lame ass R5000. Add in Windows NT, Glint and Voodoo offering the first low cost 3D GPU's, Microsoft buying Softimage and porting it to this platform and you have the inflection point where PC's started burying RISC workstations and it is the point that SGI and SUN started a long decline in the workstation market.
On the subject of Itanic that was no doubt a key contributor to Intel's current problems in the CPU market. They squandered far to much of their R&D budget on these processors which only work well on supercomputing apps, and were in general a disaster of EPIC proportions, while AMD did 64 bit sensibly and in a way that sells to a mass market.
I wager Intel is going to disappear out of the supercomputing, server, workstation and desktop market, and be forced to survive on ARM and Centrino in devices(phones and settop) and laptops. That's not necessarily a bad thing since it appears that is where much of the future of computing is heading.
@de_machina
I can't believe nobody has mentioned iLife. To me, it's by far the biggest reason to switch to a Mac. Nothing else compares in the quality of features and integration for creating, managing, and sharing your own photos, home movies (editing and creating DVDs), music (recording your own or listening to stuff you bought), and now websites. There's other software out there, but show me one package that does all of what iLife does, as easily, seamlessly, and perhaps most importantly, for so cheap!
Screw the OS holy wars, the price flamewars, and the hardware bitchfests. iLife is why anyone with a creative bone in their body should check out a Mac.
I can't imagine it's worth it to Apple to maintain classic emulation. Really, it's a small demographic who use it today-- and probably all of them write for lowendmac.com. Sure there will be someone who can come up with a "but I still use Microsoft 5" or whatever, but they're exceptions, not the vast majority. That might suck for a few people, but it's just not enough to matter.
If there's a classic app you or someone else needs to use so badly, consider that you can pick up a 400mhz G4 tower or a G3 Powerbook for less than $200 (possibly less than $100) and run OS9 natively.
So if your new Mac offers a 2.1Ghz CPU and a new Dell has a 3.0Ghz of the same product type - it's clear. The Dell is a lot more powerful. And the general public understands that.
Even that's not quite true. The power of a cpu is what you can do with it, not its clock-speed. A faster chip of the exact same line is not more powerful if the software is less powerful.
For example, for day-to-day tasks, a slower Mac is more powerful than a faster PC. For games, a slower PC is much more powerful than a faster Mac.
When it comes to iLife style apps, a 1.25GHz G4 Mac is far more powerful than a PC (Windows or Linux) of any speed.
Or, put another way, what's more powerful: running Windows Movie Maker on a 3 GHz cpu, or iMovie on a 2 GHz cpu?