Slashdot Mirror
Speculation on Real Reasons Behind Apple Switch
/ASCII writes "There is an article over at Ars Technica with some insider information about the reasons behind Apples x86 switch, given that the new IBM processors seem to be a perfect fit for Apple. The article claims that Apple hopes to power its entire line, from Servers to desktops to iPods and other gadgets with Intel CPUs, and that by doing so, they will gain the same kinds of discounts that Dell get."
Inside the big switch: the iPod and the future of Apple Computer
By Jon "Hannibal" Stokes
Sunday, July 10, 2005
If you've been following the Apple-to-Intel transition, you're going to want to read this whole article. Why? Because I'm going to do something that I almost never do: spill insider information from unnamed sources that I can confirm are in a position to know the score. Note that this isn't the start of some kind of new trend for me. It's just that all this information that I've been sitting on is about to become dated, so it's time to get it out there.
As I said in my previous post on the 970MP and FX unveiling, the new PowerPC processor announcements from IBM raise a number of questions about timing, like, when will these parts be available? how long has IBM been sitting on them? why the apparently sudden leap in performance per watt on the same process after a year with so little improvement?
The announcements also raise serious questions about why, if these great parts were just around the bend, did Apple really jump ship for Intel? Was it performance, or performance per watt, as Jobs claimed in his keynote speech, or were there other, unmentioned factors at work?
I have some answers to those questions, and I'll pass them along below. However, those answers come complete with their own vested interests, so feel free to interpret them as you will.
First, let's talk about the broken 3GHz promise. It's apparent in hindsight that 3GHz on the 970 was never going to happen on a 90nm process without lengthening the 970's pipeline, which is a fairly significant change. Who knows why IBM promised Jobs 3GHz? All we know is that IBM tried to hit that target without the needed pipeline change, and missed it.
The laptop G5, which is the long-rumored and now-announced 970FX, has supposedly been ready to go into an Apple laptop since at least early last month. And for what it's worth, yes, Apple was offered the Cell and other game console-derived chips. In fact, IBM routinely discloses its entire PowerPC road map to Apple, so pretty much anything PPC that IBM puts out is not only not a surprise to Apple, but it's potentially available for Apple's use.
So why didn't Apple take any of these offers? Was it performance, as Jobs claimed in his keynote? Here's something that may blow your mind. When Apple compiles OS X on the 970, they use -Os. That's right: they optimize for size, not for performance. So even though Apple talked a lot of smack about having a first-class 64-bit RISC workstation chip under the hood of their towers, in the end they were more concerned about OS X's bulging memory requirements than they were about The Snappy(TM).
One of the major factors in the switch was something that's often been discussed here at Ars and elsewhere: Apple's mercurial and high-handed relationship with its chip suppliers. I've been told that the following user post on Groklaw is a fairly accurate reflection of the bind that Apple put itself in with IBM:
I've worked with Apple
Authored by: overshoot on Sunday, June 12 2005 @ 08:56 PM EDT
and I can tell you, there's a very good chance that they outsmarted themselves into a "no bid" response from IBM.
Part of Apple's longstanding complaint against IBM was that Apple would announce a new computer with a new IBM processor, sales would skyrocket, and IBM wouldn't have adequate supply. We've all heard the story. Here's my take:
Apple negitiate for a new processor chip. Being Apple, they want "most favored customer" treatment, with fab-fill margins for the vendor. What's more, they want this for what amounts to a custom processor chip, so any oversupply will just sit on the shelf until Apple decides they want them, and sometimes Apple will let them sit a while to see if they can get a price break -- it always pays to remind the world that one is, after all, the Steve Jobs.
With terms like that, custom chip vendors only start as many lots as the customer contracts to accept
They claim -Os is to remove bloat, not increase performance :-) Thing is, for kernel type code the resulting code is actually _faster_ than with gcc -O2, since there is a lot less cache pressure.
The Fedora kernel people have benchmarked this quite a bit (and now compile kernels with -Os too), the difference is quite measurable, 5%:ish in some benchmarks.
Compiling with "-Os" (optimize for smaller code size) is not always at odds with speed, as is implied in the article.
While for some trivial benchmark code -O4 may generate faster code, for real-world applications keeping your code in cache is worth more than loop unrolling - so in real-world stuff often -Os is better than -O[2345].
www.eFax.com are spammers
No. Volume discounts are not illegal. Only offering volume discounts to customers who stay away from your competitors products might be, but that is not the same thing.
If bulk discounts where illegal, Wallmarts would be out of buisness and everyone would have to shop at 7-11.
Try out fish, the friendly interactive shell.
No he doesn't. In fact he expicitly states that in the article:
"If you read all the way to the bottom of this article, and you think that my basic thesis is that "the Mac is doomed and Apple is planning to quit selling personal computers," or some other such fatuous nonsense, then you need serious remedial help with reading comprehension. I already made my point about Apple's shift in focus from the desktop PC (as exemplefied by the Mac) to the post-PC gadget (as exemplefied by the iPod) in a previous article, where it didn't occasion nearly this much contention and idiotic, defensive ranting."
IIRC i read that the new iPod will be moving to use 2 ARM9 processors, instead of the current set up(2xARM7 and a ARM9). ARM9 processors are generaly 'quicker' than XScale processors, not necessarily more MHz but like for like, ARM performs better! (trust me i work for ARM)!!!
.
You, sir, seem to be suffering from a serious humor deficiency.
My recommendation: Go read America's finest news source until you begin to laugh again. Then come back here.
____
~ |rip/\/\aster /\/\onkey
I'm not sure what you mean by this. You should read up on the Church-Turing thesis. Basically, it can be proven that a very simple instruction set (I think the minimum is 3 instructions[1]) can run any algorithm. The question then becomes, what instructions should be implemented as a single instruction on the chip, and which ones should be implemented as a combination of instructions. Generally, it turns out, it is a good idea if all of your instructions take the same length of time to execute - this makes interleaving different instructions much easier. It therefore makes sense to have a relatively simple instruction set.
The trend towards CISC ended with things like the VAX. Back when people used to program in assembly, it made sense to have complex instruction sets to make things easier for the programmer. The VAX included things like an evaluate polynomial instruction, for example. Of course, this was quite unwieldy, and so a lot of the instructions were implemented as microcode - you they were automatically translated to a set of simpler instructions.
With the development of high-level languages, it emerged that compiler writers were not using these complex instructions, they were implementing them directly in simpler instruction. It then made more sense to focus on making a small set of instructions run quickly (which, it turns out, is easier and therefore cheaper).
Note that `CISC' chips are not really CISC anymore. They do the same `emulation' that RISC chips do. When you run x86 code on a Pentium each instruction is broken down into simpler instructions and then these are executed on the RISC core. The Pentium 4 (and, I believe, the Pentium M) cache these micro-instructions, so they don't have to do the translation twice.
[1] Zero, Increment, and Conditional Jump, for example - try it, you can do addition simply, multiplication by repeated addition, then build more complex algorithms from there.
I am TheRaven on Soylent News
"We are looking at using some chips from AMD."
It's a good point, but for the multimedia stuff that Mac does best, MHz still matters.
For example, any Anandtech P4/eMT64 vs. AMD64 comparison will show that AMD64 is faster at everything except audio & video compression (even with the new A64's that have SSE3).
"I don't know, therefore Aliens" Wafflebox1
It's been obvious for a while that the "real reason" is that Apple's needs are met better by Intel's roadmap than IBM's, and Apple doesn't have enough marketshare to make it worthwhile to change that.
Yes, the new 970FX chips are an improvement over the current tech. On the other hand, it's not mind-blowing compared to Intel's current line-up, much less what's in the pipeline. I'm supposed to be impressed by an announced 13W @ 1.4GHz and 16W @ 1.6GHz when Intel has been selling 10W @ 1.5GHz for months?
Even the dual-core Yonah core, slated for volume production first quarter of 2006, is quoted as staying within a 25W envelope @ 2.13GHz. Speeds for the low voltage, ultra low voltage, and single core parts aren't released yet, but Intel has made it clear that it's aggressively pursuing lower power designs and that notebooks based on the next generation of chips will "use approximately 33% less power".
ARM is an instruction set, not a processor. The XScale is an ARM processor.
RISC is a type of instruction set, of which the ARM instruction set is one. ARM also refers to the company that created it, to several of the processors they themselves produced (e.g. ARM 2, ARM 3, ARM 600, StrongARM) and to the cores they sell to other people.
I think it cuts both ways. W/ the PPC processor the idea of performance could always be a bit vague. Yes you were paying $2k to $3k for a desktop but because the G5 ran faster on some codes and/on some benchmarks it could be justified. Now that they have switched to intel, they no longer can make the high profit margins. People are willing to pay $100->$200 (maybe) a premium for same performance for a better OS, but the $400->$500 margins are no longer going to work. Dell sells platforms for btwn $300->$1500 for the top of a line, If apple goes much beyond that the obvious question is why pay X extra for an OS that doesn't even run the equivalent amount of software (from a user perspective). Oddly, they seem to be going the route that low end SGI/HP/IBM UNIXers have gone. Intel/AMD/Opteron single or dual processor chips w/ a premium sticker attached to it which comes w/ a Unix that is either the same as what everyone has (Linux) or worse.
So what's this about "any" P4 vs A64 tests that show that P4s are superior in audio & video compression?
Condemnant quod non intellegunt.
i guess you need to learn grammar and how to read title case. Oh wait, I'm sure that confused you with a lower-case 'i'. Was that a proper pronoun? Grow up; (OMG it's a scary semi-colon) if you can't understand title case, then obviously you won't be able to comprehend an article. Very basic grammar rules here, nothing "challenging".
(mod me down, i know you will. just be sure to drag the trollz down with me.)
You can purchase a spare parts kit. In essence, it's an extra copy of all the components that could possibly fail. Between this and a call to tech support, I could have a server running in a few minutes. If I use a part, they ship me a new one.
Dell doesn't even sell spare logic boards. The last time we called for hardware support, the instructions they gave would have erased our RAID array. This for a simple diagnostic. The mac equipment will TELL you what is wrong on a happy JAVA app.
"Learning is not compulsory... neither is survival."
--Dr.W.Edwards Deming
The PPC is just a plain better processor. The x86 has done well, and I'm impressed how far it has come from a chip that originally wasn't powerful enough to be a calculator. The PPC was designed recently, and there fore suffers fewer legacy issues. I mean, Pentium 4's still have real mode, for example.
The Pentium 4's design is superior in the ranks of a server chip, but I'm afraid they simply over-designed it to the point they're carrying too much baggage in the means of pipeline stages to keep encoders happy. And I really hate AMD users who validate everything by the benchmarks since almost any benchmark currently available is optimized either which way, and the only way to really tell is to use the damned hardware and get a feel for it.
Of course, it may seem trollish to bag at benchmarks, but really, I simply don't feel like they prove anything anymore. None really test for what they need to (multiple instructions? multiple threads? Fully qualified float, integer and vector math? Please, 3D mark used to be good, but they've been caught cheating, so let's not go there.
I really love and admire both AMD and Intel (and these days I'm really more on AMD's side, simply because I'm rooting for the underdog, and I'm a poor college boy), but I feel like neither of them are playing fair anymore. AMD gets a rush off copying every implementation Intel's ever made, and yet, when Intel does the same they cry foul? That's not sportsmanly at all, that's just being a whiner. AMD's still a multibillion dollar corporation, and they CAN catch Intel; they just need to do it and shut their upper lip about it.
Intel on the other hand, still employs a great deal of engineers that sit on their asses because of their current marketing team. There is so much they could be doing with Pentium M that they simply won't do, hopefully this deal with Apple will help them to. There's so much they could do with BIOS that they just won't do, hopefully Apple can alleviate this as well. There is simply so much innovation to be had, and it's so stifled by old technology, bad marketing and biggotry.
On a side note; if Intel went with the Pentium M now, that could really be a sign of monopolistic behavior that AMD could use. The Pentium M was designed by nature to run cooler, period, and thus, can outclock its competitors. I believe the IPC is still a bit lower than A64, but mainly due to A64's bitwidth. We'll see in about a year.
"Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
The G5 was a dandy machine, and our customers couldn't get enough of them. The trouble is, neither could Apple!
Apple was leaving a pile of money behind, every single quarter that they had to put up with IBM's supply limitations. I didn't have access to the figures, but I would estimate that sticking with IBM was costing Apple upwards of a billion in revenue per quarter.
One thing that Apple knows they're going to get from Intel, is reliable supply of all the CPUs they can use.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
I've been repeating this every time FUD about "the switch" comes up:
Apple announced a two-year long transition from PowerPC to Intel/x86 chips. Two years.
That's two years that at least some of Apple's computers will contain PowerPC chips. IBM just announced some tasty new PowerPC chips. Doesn't anyone else think that *maybe*, Apple is going to be buying some of them in the next two years?
Don't become a regular here -- you will become retarded.