This discussion has been archived.
No new comments can be posted.
Reports Say Apple May Manufacture Its Own Chips
Related Links Top of the: day, week, month.
There is no opinion so absurd that some philosopher will not express it. -- Marcus Tullius Cicero, "Ad familiares"
design it's own chips, not manufacture (Score:5, Informative)
Re: (Score:2, Funny)
Re: (Score:2)
I don't think anybody has seriously suggested that Apple is planning to build their own fab.
According to the article they bought PA Semi who has a fab...
Big deal (Score:2)
People have been designing Apple chips [blogspot.com] for a long time. Wake me when they make an edible iPod shuffle.
Manufacture or design? (Score:5, Informative)
There's a big difference between manufacturing a chip and designing one. Unless Apple suddenly acquires the capital and know how to run a fab, manufacturing is best left to foundries like TSMC.
I'd even be surprised if they did the design completely in-house. Most likely it would be a collaborative effort with an already established low-power design house like ARM.
Re: (Score:3, Insightful)
Course, people were saying the same thing about manufacturing laptop cases, like out of single blocks of aluminum.
Steve jobs seems to get a hard on from fully automated factories. The NEXT factory could produce thousands of computers a week, with a handful of employees.. (they could just never sell that many)
Re: (Score:3, Insightful)
I'd guess there's at least a few billion dollars difference between a reasonably up-to-date fab and the people/infrastructure it requires, and what is required to cast and CNC chunks of metal (unless it's something like sub propellers). If Apple was throwing around that kind of cash it wouldn't be a secret.
Re: (Score:3, Informative)
You mean "an already established low-power design house like" PAsemi http://www.forbes.com/2008/04/23/apple-buys-pasemi-tech-ebiz-cz_eb_0422apple.html [forbes.com]
Re:Manufacture or design? (Score:4, Interesting)
Re: (Score:3, Interesting)
Yes, if only Apple bought out a chip design company [slashdot.org]. Then maybe they could design their own chips.
Re: (Score:2)
I agree on the fab side unless they have a few extra billion laying around, but they could go hire people with chip design experience and move that in house.
Re: (Score:2)
Not necessarily. These days, when you make a CPU for some type of embedded application (like a mobile device), you don't design an entirely new CPU core. You just take an off-the-shelf core, like an ARM or PPC, and then create a SoC (system on chip) using that. The core stays the same, whether it's an ARM9 or ARM11 or PPC e300 or PPC e500, but all the other stuff on the chip is different, and specific to its application. If it's for something like an iPhone, it probably has a logic block that controls t
In other reports... (Score:5, Funny)
Re: (Score:3, Funny)
Re:In other reports... (Score:5, Funny)
It was actually revealed that their real motivation behind the Apple team's efforts is to build an uber sophisticated intelligent computer system capable of downloading Steve Jobs' brain in case he becomes too ill to continue his role as RDF overlor...er...CEO.
I, for one, welcome our new mock-turtlenecked overlord.
Re: (Score:2)
Rather have that then Gates or Ellison.
Apple Product Cycle revisions (Score:4, Funny)
Looks like the Apple Product Cycle [misterbg.org] may have to be revised slightly.
it seems that this will be a variant of the ARM (Score:5, Interesting)
Re:it seems that this will be a variant of the ARM (Score:5, Funny)
Yes, at last Apple will step out from under Dell's shadow!
Re: (Score:2)
And that's why they dropped PowerPC in favor of Intel's chips? I don't think so.
Re:it seems that this will be a variant of the ARM (Score:4, Interesting)
Comment removed (Score:4, Insightful)
Re: (Score:2)
They switched to Intel instead of AMD because they had had quite enough of vendor disappointments. AMD was a far riskier prospect.
AMD was not a riskier prospect, AMD was a backup plan while intel was the lowest bidder. Or, so I would speculate. Remember, this was a time when it had become clear that AMD could be taken seriously. A friend of mine has an Athlon 64 laptop from back when that was a new idea; it's still one of her best machines. (I helped her pick it out, at Best Buy... on clearance. But I wouldn't expect many savvy customers there anyway. IIRC it was around $900 with 1GB memory and maybe 60 or 80GB disk and I think the po
Re: (Score:2)
Re: (Score:2)
They also don't seem to be able to stuff a quad-core into a Macbook...
Re: (Score:2)
Re: (Score:2)
Apple dropped IBM's POWER series for Intel marketing dollars and Intel x86 familiar developers. The Power series is faster than the Intel or AMD chips, and has been for years. IBM released a 4.7GHZ dual core chip more than 2 years ago...
And this fits in a low-power laptop like the MacBook Air?
Re: (Score:2)
THIS is what it takes to get flash running on ARM?
Sheesh. Next thing, you'll be telling there's an IWM in the next iMac... harrr....
Obligatory (Score:2)
Should we even be falling for this stuff anymore?
Apple chips (Score:4, Funny)
What kind of dip goes with that?
Re:Apple chips (Score:5, Funny)
This [wikipedia.org] kind, of course.
Re: (Score:2)
dip?
I doubt Apple uses DIP at all. I would expect something more like SOIC.
May? MAY??? (Score:5, Informative)
" ... Reports Say Apple May Manufacture Its Own Chips ..."
" ... "PA Semi is going to do system-on-chips for iPhones and iPods," Apple CEO Steve Jobs said, according to The New York Times during Apple's June 2008 Worldwide Developers Conference. ..."
From the Horse's Mouth, 9 months ago, announced publicly at the WDC. I think I would be going with " ... will manufacture it's own chips ..." since that's what they said they would be doing, right out loud in front of God and everybody.
Scattered reports??? (Score:2)
Apple's purchase of PASemi was big news. Scattered reports, my ass.
Re: (Score:2)
Shouldn't surprise, they own a chipmaker (Score:5, Insightful)
Re: (Score:3, Insightful)
They're a chip designer. Chip makers actually have fabrication plants.
Re: (Score:2)
Re:Shouldn't surprise, they own a chipmaker (Score:4, Insightful)
Maybe, but it's a line not often drawn by many ... ATI never had a fab.
drawing lines (Score:2)
Title correction... (Score:3, Interesting)
Reports Say Apple May DESIGN Its Own Chips
The objective likely to be more proprietary enhancements [engadget.com] to their product lines that require licensing and royalties from secondary vendors who wish to manufacture and sell peripherals and products to work with Apple products. Its all about building monopolies, U.S. businesses believe competition is a bad thing.
Re: (Score:2)
The auto industry shows why a lack of competition and high barriers to entry are bad. When presented with serious competition, like the Tucker, the US automakers resorted to bribing politicians and judges to shut down their smaller competitors. When they couldn't stay solvent themselves, they got sweet government loans and bailouts (not just recently, but in the 80s too) that other companies didn't have access to. Despite all this history, our stupid government continues to bail out these fumbling, faili
Apple chips? (Score:3, Funny)
Re:Apple chips? (Score:4, Funny)
I know for a fact that those chips are from a different company! Do Apple's lawyers know about this?
Apple Rumors (Score:2)
Not newsworthy. Apple fanboys play telephone [wikipedia.org] more creatively than anybody this side of John Dvorak.
I always wished... (Score:2)
They had bought Transmeta (or, more recently, Sun) - Both have world class low-power technology. At least Oracle will make good data centre use of that. (I hope.)
I like Apple chips. (Score:2)
Re: (Score:3, Insightful)
Re:It didn't work for microsoft... (Score:5, Insightful)
So the difference is that neither one of them really innovates? I don't see the difference. Sure, Apple is good at repackaging things to be pretty and easy-to-use, but that doesn't matter when it comes to chips. In this case, they will *have* to innovate to turn their investments into something useful.
I think this move has more to do with Apple's obsession with controlling everything - they'd like to be a vertical company. It's a risky move, because hardware is a costly industry to enter. Will their recent purchases be worth it? Very possibly, it's an interesting gamble.
innovation, custom chips == !hackintosh (Score:5, Insightful)
Steve Jobs said it well: "Real artists ship."
It's a very entrepreneurial idea -- quit all the talking and hand-waving and actually ship something! There's not much value in developing great ideas that never get out of the lab. :) I would say MSFT is far better at marketing their ideas and capturing market share, while Apple is better at inventing. Others will have a different view.
As for the claim that neither innovates? Hogwash. Taking an idea and integrating it into a viable product IS innovation by definition -- it is something that has not been done before that point. Both MS and Apple innovate, to different degrees, which we can squabble about, ad infinitum.
But back to the original subject, I suspect Apple's desire for custom chips comes not from a desire to save power (there are already many viable low-power CPUs and chipsets available) but rather a desire to fight off Hackintosh clones (OSX running on non-apple hardware, such as the Dell mini 9 or generic desktop PCs). Technologically, there's no reason why this can't happen but one must consider that Apple's hardware sales are quite profitable and that share is worth protecting.
Re:innovation, custom chips == !hackintosh (Score:5, Insightful)
Re: (Score:2)
But back to the original subject, I suspect Apple's desire for custom chips comes not from a desire to save power (there are already many viable low-power CPUs and chipsets available) but rather a desire to fight off Hackintosh clones (OSX running on non-apple hardware, such as the Dell mini 9 or generic desktop PCs).
This doesn't make sense to me for their desktop machines. If that's their plan, they have to either 1) have all of the standard chipset for a commodity PC plus their proprietary chips and duplicate functionality, or 2) lose Windows compatibility because some subset of the operations are being pushed onto proprietary chips, and Windows doesn't speak their instruction set. Option 1 would increase the complexity on their motherboards and increase power usage, and I'd bet pretty strongly that they're not goi
Re:innovation, custom chips == !hackintosh (Score:4, Funny)
I can see it now:
Apple Strategic Planning Meeting
Steve Jobs: "So, from this point on, we'll build our own iProcessors, "
Edward Applebee: "But sir, then our macs won't be able to run windows!"
Steve Jobs: "Hey... you next to Edward, I want you to go over and hit that fucker in the face, really fucking hard!"
Re: (Score:3, Interesting)
I doubt that the hackintoshes are a serious concern to Apple. The fact they exists demonstrates that a certain market exists, but not one that's likely too profitable right now. It's something relatively few people are willing to do, or even have the knowledge (or at least patience) to pull off. Not much of a problem among the Slashdot crowd, but certainly among the general public. More importantly, in order to fight that off, Apple would have to transition back off of the x86 architecture - not a feat o
Re: (Score:2)
I would say MSFT is far better at marketing their ideas and capturing market share, while Apple is better at inventing.
Considering I've never met anyone who thought the MSN butterfly commercials were anything besides ghastly, I'd have to say that Apple is far better at marketing. Unfortunately, despite their clever ads, Apple hasn't been able to capture much marketshare from MSFT simply because of their monopoly power (and probably other factors, like lower initial cost for a Dell/HP PC vs. an Apple one).
W
Re: (Score:2)
I think this move has more to do with Apple's obsession with controlling everything - they'd like to be a vertical company. It's a risky move, because hardware is a costly industry to enter. Will their recent purchases be worth it? Very possibly, it's an interesting gamble.
This worked really well for Commodore - and they kinda died over it. Up until their bankruptcy almost all Commodore machines used custom chips designed and manufactured by them.
While initially it gave them an edge over the competition (si
Re:It didn't work for microsoft... (Score:4, Insightful)
You are completely mistaken.
Companies like Foxconn and ASUS build Apple's hardware.
Re:It didn't work for microsoft... (Score:5, Insightful)
Companies like Foxconn and ASUS build Apple's hardware to Apple's specifications.
fixed it for you.
Re:It didn't work for microsoft... (Score:4, Insightful)
Let me fix that for you:
Companies like Foxconn and ASUS build Apple's hardware to Apple's specifications, as they do for Dell and just like they used to do for Packard Bell.
Re:It didn't work for microsoft... (Score:5, Insightful)
I believe the original point was that microsoft has never attempted any serious hardware development; so comparing microsoft's supposed failure to design "simple hardware" to Apple's attempt to design "real hardware" is stupid.
Generally the hardware is designed well by every company; it's the software where things fall down. I have several Apple and Microsoft Keyboards and Mice.
Of my peripherals that are at least 2yrs old that should still be supported:
1xUSB MS mouse = support officially discontinued(3 out of 5 buttons work with default driver).
1xUSB Apple mouse = supported (but only 1 has button)
2xUSB Apple Keyboards = supported (but new Macs/PCs no longer support the power-button on the keyboard to power on when turned off)
All in all, a pretty pathetic amount of support. Microsoft drops support for their own USB mice(you can still find 3rd party drivers to enable all 5 buttons). Apple didn't officially drop support, but no longer provides the needed circuitry on their motherboards to power-up a computer via a USB keyboard's power button(I'm wondering if this is so they use less power when turned off).
Re:It didn't work for microsoft... (Score:5, Insightful)
Also, didn't MS design that one thing, what was it called, the xbox (or something)? I heard it was just a repackaged computer.
Re:It didn't work for microsoft... (Score:5, Informative)
Re: (Score:2)
Which is still one step closer to Apple than it is to Microsoft.
Apple is like Dell. Microsoft licenses software to Dell. That's the difference.
Not that they don't invent anything themselves. Who else has a magneticly-attached power cord? (If you know of one, please let me know -- it's a very cool idea, but I don't want to buy a Mac.)
Re: (Score:2, Insightful)
Let me fix that for you:
Companies like Foxconn and ASUS build Apple's hardware to Apple's specifications, as they do for Dell and just like they used to do for Packard Bell.
I'll fix it for everyone. Apple builds them, tests them and sends them to be assembled to their specs. Done. Get over it.
Re: (Score:3, Interesting)
Let me fix that for you:
Companies like Foxconn and ASUS build Apple's hardware to Apple's specifications, as they do for Dell and just like they used to do for Packard Bell.
One more time and we'll get it right: Companies like Foxconn and ASUS build Apple's hardware to Apple's specifications. Companies like Foxconn and ASUS also build hardware for Dell and Packard Bell, most of which was (reference) designed by engineers at Intel. Hell, for the longest time, Dell's boards had "Intel" silkscreened on them, since they were literally carbon copies of Intel's reference board and nobdoy bothered to change it.
Re: (Score:3, Informative)
Re: (Score:2)
Don't mistake "build" for "design."
Apple design their hardware, inside and out, down to the motherboard and chip placement. They outsource the manufacturing, but the actual system builder isn't just sticking an Apple badge on a generic piece of hardware.
These aren't just Intel reference boards being stuck into Macs.
Re:It didn't work for microsoft... (Score:5, Interesting)
Nope, not even close. Apple *designs* or works with manufacturers to create custom *designed* boards and hardware but they build nothing. They are the same chips and chipsets as Dell, which actually does the same thing and custom *designs* their gear just like Apple.
Take off the rose colored glasses please.
Re:It didn't work for microsoft... (Score:5, Informative)
Before the Intel switch, Apple absolutely designed its own chipsets and boards. Apple was responsible, for example, for the first marrying of the PPC 970 and HyperTransport.
Apple has never owned a fab, but then, neither do many dedicated chip "manufacturers."
Re: (Score:2)
According to the article they own P.A Semi...
Re:It didn't work for microsoft... (Score:5, Informative)
I don't know for sure, but just because they own a chip "manufacturer" doesn't mean they own a fab. I'd be willing to bet they don't. There's lots of semiconductor companies that don't have fabs at all; they're called "fabless". P.A. Semi was probably one of them. Here's some others you may have heard of: Qualcomm, Broadcom, NVIDIA, Marvell, MediaTek, ATI (before AMD acquired them), Xilinx. Here's an article [wikipedia.org] about them. These companies simply design chips; they get other companies called "foundries" to make their chips for them. The largest and oldest of these is TSMC, a Taiwanese company.
Re: (Score:3, Insightful)
Nope, not even close. Apple *designs* or works with manufacturers to create custom *designed* boards and hardware but they build nothing. They are the same chips and chipsets as Dell, which actually does the same thing and custom *designs* their gear just like Apple.
So in the same way, all cars and buildings are exactly alike because they all use exactly the same materials, the design is just a bit different. Right?
Yeah, the CPU and chipset are the same. They even use copper and solder to make their boards, oh my! But there's are millions of ways to lay out the same components, and they do yield different results. Take cities: they all have similar needs (hospitals, police stations.. just play Sim City), but how you lay them out can be the entire difference between
Re:It didn't work for microsoft... (Score:5, Insightful)
Just think how much [Apple] could power reduce and cost reduce if they dictated the chip-specs!
(I am an electronics engineer, and make chips that Apple buys for many of their products)
They already do, to an extent.
Chip design companies are constantly battling it out to get design wins at big companies like Apple. If Apple tells them, "Hey, we want this chip to do X and Y while consuming Z mA," then those companies are going to try their best to meet those requirements so that they can get Apple as a customer.
Your assertion that chips are being overcomplicated for the purpose of driving up cost is incorrect. Semiconductor companies are constantly trying to simplify their chips' designs, in order to improve yield and reduce costs, while charging the same price to their customers. It's much, much easier to improve margin than it is to convince your customers to pay more.
I doubt that Apple will be able to substantially improve cost or power consumption. While they do have some experience in chip design, it's highly unlikely that they'll be able to go in and do a better job than all of the companies that do nothing else.
Re: (Score:3, Informative)
Microsoft doesn't have that experience, because they don't build systems.
So the Xbox, Xbox 360 and Zune are what now?
Re: (Score:3, Insightful)
Re: (Score:2)
Apple is not going to waste money developing their own chips just for bragging rights.
That's right, they won't do it just for bragging rights. They'll do it for a compelling performance, power consumption, and/or cost advantage. Right now, they pay Intel, Nvidia, and AMD a hell of a lot of money for CPUs and GPUs, and I'm sure they'll do their homework before making the next build or buy decision.
The question is, would Apple seriously get into developing their own CPUs? I find that hard to believe, if only because- in terms of worldwide market share- the Mac is still probably quite small compared to Intel and even AMD based PCs.
Given how competitive the CPU market is and how hard AMD have to work to even compete with Intel, I find it hard to believe that Apple could afford to develop a chip that was competitive in terms of price *and* performance with either Intel or AMD solely for their own use.
Comment removed (Score:4, Interesting)
Re: (Score:2)
Re:It didn't work for microsoft... (Score:5, Informative)
Apple also participated in the design of the initial ARM processors. That seems to be going pretty well. (Direct descendants of the design are in iPhone).
Apple is also a participant in LLVM, which is going to help Apple shorten the design-to-deployment cycle for new silicon.
It's going to work out just fine.
Re:It didn't work for microsoft... (Score:5, Funny)
Mini was zippy? My friend, in a Mac world the correct term is "snappy". Amateur!
Re: (Score:3, Funny)
Maybe he hadn't updated Safari yet...
Re: (Score:2)
Also, Apple has had several internal projects that were oriented towards building chips in the past. The story has it that once Steve Jobs met Seymour Cray and told him that he was using a Cray-1 to design his next computer. Seymour says, "That's great, Steve. I'm using an Apple to design my next computer."
Re: (Score:2)
Apple participated in the design of the PowerPC.
The key word here are surely "participants in design." No capital investment. No long term commitments.
In a deep recession the iPhone can be seen for what it is - a high-tech gadget in a market that is moving towards the low-tech Jitterbug. [jitterbug.com]
The basic cell phone the geek always says he wants but never actually buys.
Re: (Score:2)
If you had said the Nokia 1100, I might have not said you were insane.
But the Jitterbug? Seriously? For starters, it has a color screen - something that the Slashdot favorite "basic phone" doesn't have at all. And, second, it costs more than some modern smartphones. Yes, I know, it's not on contract... but the prepaid prices are ridiculous, so if you actually use this "basic phone," you're paying OUT THE ASS.
Re: (Score:2, Insightful)
Apple knows how they intend to use the chip, but that is not the same as knowing how the chip gets it done. Participating by writing various specifications and testing is a very long way from designing logic circuity.
That's not to say they can't do it, or wont be good at it. But your making a leap.
Comment removed (Score:5, Interesting)
Re: (Score:3, Interesting)
There's no way to cover the design costs off the miniscule number of chips compared to Intel and AMD.
On the other hand, AMD's market cap is $2.9B. Apple could buy them. Then the economy of scale would work out, wouldnt it?
Re:It didn't work for microsoft... (Score:4, Informative)
Apple also participated in the design of the initial ARM processors. That seems to be going pretty well. (Direct descendants of the design are in iPhone).
Nitpick: Acorn, not Apple, solely did the design of the initial ARM1, ARM2, and ARM3. They then spun the ARM CPU (which originally stood for Acorn RISC Machine) off into another company, Advanced RISC Machines, which was a joint venture between themselves (40%,) VLSI (who did most manufacturing of ARM CPUs and chipsets at that point - 40%,) and Apple (20%,) as Apple had expressed interest in using the chip, but didn't want to use a competitor's chip (Acorn directly competed with Apple in the personal computer market, especially in schools.)
Only the ARM6 (there was no ARM4 or ARM5) and newer had any Apple involvement, and I doubt anything newer than the DEC StrongARM had much of any Apple influence. (The ARM6, ARM7, and StrongARM were all used in the Newton.)
And, the ARM6 and ARM7 are essentially tweaked versions of the ARM3 with 32-bit addressing (as opposed to 26-bit on the previous ARMs,) and more cache and a slightly faster clock in the case of the ARM7. As for the StrongARM, it wasn't even designed by ARM, it was designed by Digital, to meet the ARMv4 ISA.
yawn - another ARM licence (Score:2, Interesting)
Apple Core (Score:2)
If they can call is something like "Apple Core" then they will be onto a sure winner.
Especially if they can incorporate lots of "Programming Interface Pathways" or some such jargon.
Tom...
Re: (Score:3, Insightful)
Re: (Score:2)
Although I never really understood why MS wants to penetrate entrenched markets, I thought the Xbox was turning a profit now... so that must be new R&D or the Zune totally bombing.
2008 Q1
http://www.joystiq.com/2008/01/24/the-xbox-turns-a-profit/ [joystiq.com]
2008 Q2
http://www.joystiq.com/2008/04/25/the-xbox-360-turns-a-profit-again/ [joystiq.com]
Re: (Score:2)
Re: (Score:3, Funny)
Why don't we list other things that didn't work for Microsoft.
Who has that kind of time? An even bigger challenge is to figure out why people bought it anyway.
The problem with Cell... (Score:2)
The problem with Cell is that in a general purpose computer, threads need to talk to each other; you can't just have non-barriered pipelines you keep fed. That's OK for some types of specialized processing, but most of that type of processing is going onto general purpose GPUs these days (e.g. OpenCL), rather than building specialized hardware for it.
A personally would have liked to work on a port of xnu to the Cell architecture, since I think it would have yielded a lot of useful information about the OS
Re: (Score:2)
You are mistaken. It's not as though Apple didn't see the Cell coming. Sony shopped it to them but they simply weren't interested [nytimes.com].
And this was way back in 2005. The Cell is arguably good in it's role as the cpu for a game console / blu-ray playback device, but that doesn't mean that is the best choice for a general-purpose computing device.
Re:Apple shoulda never left IBM, Cell woulda been (Score:2)
Re: (Score:2)
Re: (Score:2)
Well, because before hipsters and "cool" kids flocked to Macs they were amazing at... Image processing and the like. Imagine a cell CPU with say 8 cores or so and CUDA with an additional ton of cores all churning out amazing shit like Pixars latest flick or handling massive RAW files and tons of filters.
See, then fine, have your commodity Intel crap for the in-the-know crowd. Ever notice how the "pro" lines are anything but since the switch to Intel? Well, probably not because actually doing the things that
Re:Apple shoulda never left IBM, Cell woulda been (Score:2)
First of all, it was new architecture, which consisted more or less of a PowerPC chip which shares some similarities to the G5 and the SPUs which the PPC part had to feed with data. Well, to truly take adavantage of the Cell architecture, all programs would have to be optimized to the Cell architecture, which is not an easy ta