I've been reading this enormous thread and I have yet to see anyone approach this situation from a total business perspective.
The whole point about Apple and x86 isn't always about performance or technical considerations. (Yes, they are important, but let's talk business for a moment). x86 migration is about opportunity. Even if IBM has initial success with the 970--there's no market for it. We've already been down this road before with the AIM alliance. Remember the "Burn baby burn" ads? I still have promo stickers that say "Snail Inside". (I can put those on my eMac now, BTW).
Supporting the release of the 970 as Apple's new "savior" chip is going to repeat a mistake that Apple has already made. IBM doesn't have the financial resources to compete with Intel. Regardless of "where" Intel and IBM say they will be at the same time technology-wise in the next few years, Intel will always come out ahead. They have the market. If Apple locks itself into PowerPC again, OS X, which Steve Jobs has slated as the "operating system for the next 15 years" is going to suffer. And in case anyone hasn't noticed, we've already been suffering.
From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs. They need a competitive OS to run these CPUs. And that OS is our OS X.
<strong>I've been reading this enormous thread and I have yet to see anyone approach this situation from a total business perspective.</strong><hr></blockquote>
That's because it's a thread about the 970, not a thread about the future of the entire platform. Although it is starting to fray now.
[quote]<strong>The whole point about Apple and x86 isn't always about performance or technical considerations.</strong><hr></blockquote>
No, it isn't. And when this discussion comes up, as it has in threads dedicated to the subject, many other factors are considered, including cost, developer support and politics. And the general consensus seems to be that a wholesale switch to x86 is a bad choice all the way around. Opterons might pop up in Apple server solutions, but that would be the likely extent of it.
I won't rehash the reasons in this thread, because this thread is about the 970. Unfortunately, UBB's search function doesn't work.
If you decide to start another one, it would go in General Discussion, because the choice impacts not only hardware, but software, corporate politics and strategy.
<strong>I've been reading this enormous thread and I have yet to see anyone approach this situation from a total business perspective.
The whole point about Apple and x86 isn't always about performance or technical considerations. (Yes, they are important, but let's talk business for a moment). x86 migration is about opportunity. Even if IBM has initial success with the 970--there's no market for it. We've already been down this road before with the AIM alliance. Remember the "Burn baby burn" ads? I still have promo stickers that say "Snail Inside". (I can put those on my eMac now, BTW).
Supporting the release of the 970 as Apple's new "savior" chip is going to repeat a mistake that Apple has already made. IBM doesn't have the financial resources to compete with Intel. Regardless of "where" Intel and IBM say they will be at the same time technology-wise in the next few years, Intel will always come out ahead. They have the market. If Apple locks itself into PowerPC again, OS X, which Steve Jobs has slated as the "operating system for the next 15 years" is going to suffer. And in case anyone hasn't noticed, we've already been suffering.
From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs. They need a competitive OS to run these CPUs. And that OS is our OS X.</strong><hr></blockquote>
MacLuv,
Good points. However, I keep coming back to OS/2. Back in the days of windows 3.X, OS/2 was a far superior OS than windows, it ran on Intel chips, it was backed by a much larger company (IBM vs Microsoft), and it ran windows apps with little or no problems. In addition, when OS/2 was competing with windows 3.x, it had no where near the market share it has now. And, I read somewhere that IBM spent almost 4 Billion on OS/2. So, I just can?t see how putting an Intel chip inside a fancy Apple computer case running OS X is going to be any different than what IBM tried to do. OS X compared to Windows XP is not as big an improvement as was OS/2 to windows 3.x was. And keep in mind that Apple makes most of their profit on hardware. How are they going to justify selling an Apple X86 box for 20-40% more than a comparably configured Windows XP box?
"From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs. They need a competitive OS to run these CPUs. And that OS is our OS X."
-----------
Maybe a few years into the future this could turn up as a real possibility for Apple. For right now, though, apple can not afford another huge change. OS X was a big enough change that for apple to make another one so soon would cause wide spread panic among the developers. Etc., etc., etc. I'm afraid that there will not Be x86 for apple in the near future and for that matter I suspect never. X86 is old and befuddled and not something apple will base their future on.
x86 migration is about opportunity. Even if IBM has initial success with the 970--there's no market for it. We've already been down this road before with the AIM alliance. Remember the "Burn baby burn" ads? I still have promo stickers that say "Snail Inside". (I can put those on my eMac now, BTW).</strong>
If Apple moves to x86, they will die. They will not be able to compete on price with Dell and maintain a competitive operating system on their own. If Microsoft didn't have an office app monopoly, they may have chance. But the reality is that MS leverages its office monopoly and operating system monopoly far too well. Maybe if OpenOffice makes a dent, Apple can transition, but its death right now. They will not be able to compete on price for something that is not different from other box makers.
<strong>IBM doesn't have the financial resources to compete with Intel. Regardless of "where" Intel and IBM say they will be at the same time technology-wise in the next few years, Intel will always come out ahead.</strong>
Yet, AMD, a smaller company then Apple, will be able to compete with Intel?
<strong>The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs.</strong>
AMD hasn't been competing with Intel in the last 3 months, and that lag will only continue to grow in the future. AMD may be dead in year, let alone compete with Intel.
. . . Even if IBM has initial success with the 970--there's no market for it. . .
. . . From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. . .
</strong><hr></blockquote>
It sounds like you think the x86 has some special marketing appeal to it, and consumers will reject the IBM 970 because it isn't an x86 type processor. Can you tell us why you think Joe or Jane consumer will change his or her mind about the Macintosh just because it is running an x86? Really what they care about is whether it will run their software, and an x86 Mac will not do that either. People will usually switch to the Mac when it gives them something they cannot get from their present Windows PC. There is nothing magic about the brand name of the processor inside, as long as the computer does what they want.
Those who do care about technical things like the processor are interested in performance first, not so much brands and types. Consider whether this kind of discussion would even be taking place if the 970 initial clock speed was 6 GHz. You would be saying, "bye, bye x86," like everybody else. So, it is really a matter of performance, and we should pick the technology that will likely win the race, not just be ahead for the first few laps.
<strong>If you decide to start another one, it would go in General Discussion, because the choice impacts not only hardware, but software, corporate politics and strategy.</strong><hr></blockquote>
Feel free to move this post where you feel it fits.
--------------
wow, a lot to answer... :eek:
[quote]sc_markt says:
<strong>And keep in mind that Apple makes most of their profit on hardware. How are they going to justify selling an Apple X86 box for 20-40% more than a comparably configured Windows XP box? </strong><hr></blockquote>
Yes, Apple's profit design right now is based on perceived value rather than actual value. It is dependant upon people who do not yet realize PCs have become a commodoty market. It leeches off of what I call the Loyal Apple User Base (LAUB), locking us into a proprietary system that only Apple can control. Like you say, this means we pay extra for hardware that is outperformed by the competition--just because it looks pretty.
As far as selling an x86 box for 20% more... are you saying that we're paying more for the G4 processor itself? We're really paying for Apple's perceived value/goodwill. There is a great misconception that an x86 in the Apple PC would lower value, when it could actually raise it. How? That magic word Apple loves: perceived value.
As far as Apple restructuring its profit design--Apple already knows PCs are a commodity market, that's why they are setting up Apple stores. They just don't want anyone else to know the PC market is a commodity market. Steve Jobs is the man behind the curtain--and I think he's getting too comfortable back there.
For example, I don't know how much longer Apple can try and convince the public the G4 is a better choice over x86 in the consumer market, no matter how many "Genius Bars" they set up. <img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" />
-----------------------
[quote]From Algol <strong>
For right now, though, apple can not afford another huge change. OS X was a big enough change that for apple to make another one so soon would cause wide spread panic among the developers. Etc., etc., etc. I'm afraid that there will not Be x86 for apple in the near future and for that matter I suspect never. X86 is old and befuddled and not something apple will base their future on.</strong><hr></blockquote>
This is another misconception I always see from the "developers" point of view--that somehow another transition from Apple would cause a worldwide panic and engineers everywhere would rise up in disgust and burn the house down.
Unlikely.
From a business point of view, what has to be done, has to be done...
"You can handle people more successfully by enlisting their feelings than by convincing their reason." -- Paul P. Parker
Even if the migration to x86 seems unlogical from a developers' perspective, it can still be done. Everybody has a price.
----------------------
[quote]From THT<strong>
If Apple moves to x86, they will die. They will not be able to compete on price with Dell and maintain a competitive operating system on their own. </strong><hr></blockquote>
This is just not true at all. In order to convince me otherwise, you'll have to go into greater detail of why you believe Apple will not be able to compete with Dell on price or maintain a "competitive" OS. This argument always implies that Apple has seen its own shadow and should hibernate for the next fifteen years.
[quote]<strong>
If Microsoft didn't have an office app monopoly, they may have chance. But the reality is that MS leverages its office monopoly and operating system monopoly far too well. Maybe if OpenOffice makes a dent, Apple can transition, but its death right now. They will not be able to compete on price for something that is not different from other box makers.</strong><hr></blockquote>
Here I'm not clear on what you're getting at. Are you suggesting that MS Office provides the only differenciation between Dell and Apple? What about the "digital hub" Steve talks about all the time? What does this stuff have to do with the CPU inside the box?
[quote]<strong>
Yet, AMD, a smaller company then Apple, will be able to compete with Intel?</strong><hr></blockquote>
Yes, in fact, AMD has been competing with Intel for quite some time now.
As long as AMD produces CPUs, it will always compete with Intel within an oligopoly.
[quote]<strong>
AMD hasn't been competing with Intel in the last 3 months, and that lag will only continue to grow in the future. AMD may be dead in year, let alone compete with Intel. </strong><hr></blockquote>
What else do you pull out of that hat of yours? Rabbits? I'm not sure where you got this information from, but it is erroneous. As I've stated above, AMD, Intel, and any other company that produces CPUs are part of an oligopoly. As long as they all make CPUs, they are always in competition. I haven't seen any facts that AMD is pulling out of the CPU manufacturing business any time soon.
---------------
[quote] from snoopy <strong>
It sounds like you think the x86 has some special marketing appeal to it, and consumers will reject the IBM 970 because it isn't an x86 type processor. Can you tell us why you think Joe or Jane consumer will change his or her mind about the Macintosh just because it is running an x86?
</strong><hr></blockquote>
Firstly, why does one assume that when the 970 comes to market, Intel won't have something to compete with it?
Secondly, right now, the x86 kicks the G4s ass. Do you think another "Burn baby burn" revival is in order here? How long will it be this time around before Intel takes the spotlight over IBM?
[quote]<strong>
Really what they care about is whether it will run their software, and an x86 Mac will not do that either. People will usually switch to the Mac when it gives them something they cannot get from their present Windows PC. There is nothing magic about the brand name of the processor inside, as long as the computer does what they want.
</strong><hr></blockquote>
If this is the case, why don't we just slap an AMD Athlon in the iMac and call it a day?
Consumers care about a lot of things when it comes to computers--not everyone is the same. Something you don't mention is how quickly people want to use their software, which is always an issue in a commodity market.
[quote]<strong>
So, it is really a matter of performance, and we should pick the technology that will likely win the race, not just be ahead for the first few laps.
</strong><hr></blockquote>
Apple took a risk with RISC and it didn't pay off. Now you're asking me to have faith and walk down the same alley where we all got our asses kicked.
Unlikely.
IBM is in no position, and never will be, to infiltrate the market created by Microsoft and Intel. Becuase AMD manufactures a CPU that can replace Intel's, it is in a better position to compete with Apple as a partner. It will position Apple's OS right next to Windows, rather than having it in a completely different mindspace. Of course, this appears to be Steve Jobs' strategy, but things must change if Apple is to truly survive the next fifteen years.
Apple took a risk with RISC and it didn't pay off. Now you're asking me to have faith and walk down the same alley where we all got our asses kicked.
Unlikely.
IBM is in no position, and never will be, to infiltrate the market created by Microsoft and Intel. Becuase AMD manufactures a CPU that can replace Intel's, it is in a better position to compete with Apple as a partner. It will position Apple's OS right next to Windows, rather than having it in a completely different mindspace. Of course, this appears to be Steve Jobs' strategy, but things must change if Apple is to truly survive the next fifteen years.</strong><hr></blockquote>
I am not quite sure I understand the point of this very lengthy post. Put yourself in the consumer's role: you need a computer. You want to do this and that, whatever that may be, and you want a system that is going to be viable, cost-effective and do what you want. You can buy a Windows PC, a Mac, or whatever. The Mac accomplishes all the same tasks a PC does and vice versa. It's just a Mac does it in a more streamline fashion. Period. Listening to music, burning CDs, importing DV, burning DVDs, email, internet, and on and on. You don't have to worry that XP sp1 changes your drivers and now the task you did before doesn't work quite right, or hangs, or whatever. Seriously, I struggled with Windows or years, and had to reformat and rebuild my Windows PCs so many times, I just gave up on it. It is not worth my time and effort to go through all the nonsense with any Windows PC.
Thus we have the switch campaign. A brilliant move. They are not touting RISC is better in any advertising I've seen. And I paid around $2100 for my Powerbook. There's not a Windows laptop around that offers me what it does. I think it's ambiguous to argue what processor is inside a Mac or a Dell, or any other PC. It's what you can do with the computer that should make a difference, but people are so out of touch with how to really use a PC that is boggles my mind. The general public, and I hear it in conversations everyday, that they can get a P4 3.0GHz processor and blah blah blah. Who cares?! AMD, Intel, Motorola, IBM, I could give a shit, as long as I can work. And I can do that with a Mac.
BTW, have you used a new dual PowerMac? It screams! More a function of video technology that CPU at this point anyway.
Apple took a risk with RISC and it didn't pay off. Now you're asking me to have faith and walk down the same alley where we all got our asses kicked.
Unlikely.
</strong><hr></blockquote>
I can see that you truly appreciate and respect the performance that Intel and AMD have managed to get out of the x86 type processor, but your strong feelings may be affecting your willingness to use clear logic. I too am impressed with what they have done with the x86, and it is truly an outstanding technical accomplishment. I think it caught everyone by surprise. And I think, along with others, it may have begun with the performance wars, which Apple started. I can see the head honcho at Intel saying, "Get us the highest possible performance and MHz, and I don't care how much it costs." Now everyone is playing catch up. In the end, this challenge will benefit us computer users greatly.
You say Apple took a risk with RISC (good pun too) and it didn't pay off. Think about that. Is RISC the fault, or is it the company who was making processors for Apple? Really. What would have happened if it was the other way around? What if Motorola was making x86 chips for Apple and Intel had the PPC architecture? Do you think for a moment it means Apple would now have a 3 GHz chip and Intel would be struggling at 1.25 GHz? You know the outcome of that scenario. Intel would still be beating Apple in performance, but by a larger margin in my opinion.
So, nobody is saying 'walk down the same alley,' which would be sticking with Motorola. IBM will do much more with the PPC than Motorola ever could do. I'm not criticizing the talent at Motorola, as I am criticizing the commitment of management to the workstation and desktop market, or even to the semiconductor business itself. There is nothing magic about the x86 architecture, and as others have pointed out, it has a lot of shortcomings. There is a whole lot more potential in the PPC path that Apple and IBM are on, IMO.
[quote]Firstly, why does one assume that when the 970 comes to market, Intel won't have something to compete with it?<hr></blockquote>
The PowerPC, whether from IBM or Motorola or Framistan Inc., is not, in any form, on Intel's radar. AMD is, period.
The G4 may be getting old and it's potential never came to its fullest bloom (yes, blame Motorola for quite a bit of that, but also blame IBM for being block-headed about Altivec). However, the G4 was a "good enough" step up from the G3 for Apple to keep its head above water and build OS X and all those iApps. Apple isn't a Dell (And me having to point that out is ridiculous). Unlike Dell or any other box assembler, the UI & software are closely tied to the hardware and vice versa. Do you think Shake was purchased with only the G4 in mind? I don't believe so.
And lastly, don't assume Intel is pulling its punches. They decided to win the war (over AMD) with gigahertz first, performance second and sacrifice power consumption and heat. This may come back to haunt Intel in the next few years.
But again, what Intel does is not relevant to Apple.
If Apple moves to x86, they will die. They will not be able to compete on price with Dell and maintain a competitive operating system on their own. </strong><hr></blockquote>
[quote]Originally posted by MacLuv:
<strong>
This is just not true at all. In order to convince me otherwise, you'll have to go into greater detail of why you believe Apple will not be able to compete with Dell on price or maintain a "competitive" OS. This argument always implies that Apple has seen its own shadow and should hibernate for the next fifteen years. </strong><hr></blockquote>
This is amazingly simple.
Thought experiment: Assume Apple has successfully done all of the other steps they would need to do to make an x86 Mac. Answer a couple of simple questions:
Who is going to get a cheaper price on x86 chips? (Choose one: Apple, Dell) Note history of Intel disliking (for good reason) Apple.
Who is going to get a cheaper price on motherboard chipsets? (Choose one: Apple, Dell).
Repeat for all other components. The 86Mac would be cheaper (since all the components would now get mass produced at a larger scale), but they would NOT be cheaper _than_Dell_.
Dell has a 2% R&D burden, Apple has about an 8% R&D burden. On hardware. IGNORE the 'Dell doesn't make their OS' issue(s).
So Apple makes a sweet little $500 x86 Mac. Either (a) Dell can do the _exact_ same box for less, or (b) Apple puts flourishes into the design -> "perceived value" higher than "actual value" (your words).
_IF_ Apple switched, and I were Dell, I would want them DEAD. Wanting them dead is a sound business decision - a couple of years of mild hardship is a very small price to pay for a monopoly. I'll grant you that the $500 x86-Mac might be a nice box. But if I were Dell, I would sell a box no more than a couple of months after the intro of that 86Mac with precisely the same hardware - for at least $100 less. For _every_ product that Apple made that Dell could compete head-to-head on.
Apple has to either profit on the sales of these 86Macs, or gain lots of marketshare. Dell can cut the margin to 0 (there are lots of other product lines to pick up the slack). Dell can sell at a slight loss even - plenty of money stored up, and a clear advantage to killing off a competitor like Apple.
So assuming all of the other issues in the conversion to x86 cost $0, and 0 time... Apple would be making a box that is, you guessed it, costing more than the Dell. For people that enjoy using MacOSX, the extra price might be worth it, but it won't have the 'other 95%' saying "Wow, 86Macs! $500!" There wouldn't be any programs (of _any_ sort) that run faster on a 86Mac than on a Dell. When you eliminate product differentiation to compete in a DIRECT head to head competition you had better be selling more product and have a lower overhead. Neither of those will be the case for Apple in the next two years. Period.
That isn't where we were at the start of this year. There do exist real programs that have historically run faster on Macs - but those programs rely on the ppc or altivec! There are a couple of places full of dedicated PC lovers who repeatedly denounce the speed of the mac, and accumulate benchmarks to "prove" it. Fine. But some of those same sites _STILL_ have the top of the line Macs beating dual top-of-the-line Athalons in _some_ aspects of the benchmarks! Admittedly, it is in a short (and shortening) list of benchmarks that this is true, but there are places where a Mac is (or at least has recently been) the fastest machine for the job. But the 970 will at least reclaim dominance _in_the_places_Apple_needs. It might not be an overall faster processor when debuted, but things like photoshop benchmark tables are going to be re-written(even when run by PC guys). Just because the ppc970 doesn't (or won't) dominate _all_ benchmarks doesn't mean "That's a useless dead-end chip for Apple to adopt".
What I'm saying is, if I as a desktop computer purchaser needed a boatload of double precision floating point _today_, a dual athalon is the tool of choice. If I need parallelizable single precision floating point, the G4 is still up there. There are clear, objective reasons to purchase alternatives to Intel. If Apple did switch, we'd first be bombarded by a year of whining about 'Jeeze Apple's overpriced and gouging folk!', then another year of 'Why bother paying the Apple tax, this box is identical & XP is close enough'. -> spiral of doom.
Name a boxmaker or OS vending company that switched to x86 (LOTS of choices mind you) that can be considered a raging success story because of that switch.
This is just not true at all. In order to convince me otherwise, you'll have to go into greater detail of why you believe Apple will not be able to compete with Dell on price or maintain a "competitive" OS.</strong>
Dell is the largest and most agile PC manufacturer in the world, has the most inroads to consumer, business and enterprise markets, and has very good deals for all of its parts. It is well on its well to becoming a monopoly PC manufacturer. (Yes, I think HP will lose.) Now, Apple is essentially Gateway but for the fact that it has expend it's own resources for an "incompatible" operating system.
<strong>This argument always implies that Apple has seen its own shadow and should hibernate for the next fifteen years.</strong>
There may some truth in the statement. It also could be that Apple doesn't have enough business to truly innovate. It doesn't seem to have much foresight either.
<strong>Here I'm not clear on what you're getting at. Are you suggesting that MS Office provides the only differenciation between Dell and Apple? What about the "digital hub" Steve talks about all the time? What does this stuff have to do with the CPU inside the box?</strong>
MS Office is the gateway for any operating system that doesn't want to be a niche operating system. Without it, no operating system will make any inroads into the markets. Every single business uses it. Every single school uses it. The information in MS Office documents is the most valuable things we have. No one will migrate without being able to take their data with them.
The digital hub is just new marketing to sell computers. Dell can do it just as any other PC manufacturer can.
<strong>Yes, in fact, AMD has been competing with Intel for quite some time now.</strong>
The only win was the Athlon in 2001 with their 0.18 micron fab. Since Intel got to 0.13 micron first, AMD has been losing.
<strong>As long as AMD produces CPUs, it will always compete with Intel within an oligopoly.</strong>
When one manufacturer has 80% of the market, if not more, it isn't an oligopoly. Intel has all the advantages right now since they'll get to 0.09 micron and 0.065 micron first.
<strong>What else do you pull out of that hat of yours? Rabbits?</strong>
LAS VEGAS, Nov 19 (Reuters) - Advanced Micro Devices Inc. (nyse: AMD - news - people) said on Tuesday that it would embrace a strategy of developing processors for a wider range of products outside computers and called on the industry to focus on user needs rather than creating "technology for technology's sake."
...
AMD, which has fought a losing battle in recent quarters against Intel Corp. (nasdaq: INTC - news - people) for the top spot in processor speed and performance, said that making semiconductors smaller, cheaper and faster was no longer the key for an effective strategy.
...
AMD's strategy shift came on the same day that Standard & Poor's downgraded the chip maker on concerns about its ability to boost profits and weather soft demand, only hours after AMD said it would raise $300 million in a convertible note issuance.
SUNNYVALE, Calif.--Reeling from the IC downturn and competitive pressures from Intel Corp., troubled Advanced Micro Devices Inc. today announced its much-anticipated move to reduce its headcount by 15%, or approximately 2,000 positions, by the end of the second quarter of 2003.
...
Hit hard by a slowdown in its core processor business, AMD recently reported sales of $508.2 million for the third quarter of 2002, down 34% from $765.9 million in the third quarter of 2001 and by 15% from $600.3 million in the previous period.
The Sunnyvale-based company also reported a net loss of $254.2 million, or minus $0.74 a share, compared to a loss of $97.4 million, or minus $0.28, a year ago. In the previous quarter, it reported a loss of $184.9 million, or minus $0.54 a share.
AMD "Sampling" 2700+ For PCs Available In November; 2800+ Only Available To "Enthusiast" OEMs
By Mark Hachman
While Advanced Micro Devices announced its 2700+ and 2800+ Athlon XP microprocessors today, customers may be asking when they can buy the chips the company announced in August.
AMD, Sunnyvale, Calif., will launch the 2800+ Athlon XP today in "limited quantities", a phrase that analysts say hearkens back to the race to 1-GHz?when AMD, not Intel Corp., was the company that broke the tape with volume quantities of 1-GHz chips.
...
McCarron said it's possible that AMD's development of the 0.13-micron silion-on-insulator process, combined with the larger cache sizes of the delayed "Barton" processor, as well as the need to develop working samples of the Opteron and 64-bit Athlon processors, may be hindering the company. "They have a lot of balls in the air," he said.
...
"Ultimately if they can't deliver it will show up in their revenue results," Brookwood said. "One has to assume there's a method to their madness. In the case of the 2800+ the goal is cachet, exclusivity. They're only there for serious gamers." AMD's "enthusiast" OEMs will include ABS, Alienware, Falcon NW, MicronPC and Voodoo PC.
The bottom line, analysts say, is that AMD is still wagering the company on Hammer?the latest Athlons are just a side bet. Still, AMD's newfound reputation for quality may be showing signs of wear, and the fickle hearts of enthusiasts may be shifting.
"They're clearly not sustaining the level of volume (at) introduction they did a year ago," McCarron said.
Lets see, the K7 is teetering out or their 0.13 micron fab has Moto-itis, they are bleeding money like no tomorrow with 6 consecutive quarters of losses with profitability predicted 9 months away, they are showing classic signs of a company in trouble with paper launches and "strategy diversification" in the face of competition and their savior product delayed.
AMD is not in good shape right now. And if they stay this small, they aren't going to get to 0.09 and 0.065 micron fabs quick enough to compete with Intel.
<strong>I've been reading this enormous thread and I have yet to see anyone approach this situation from a total business perspective.
The whole point about Apple and x86 isn't always about performance or technical considerations. (Yes, they are important, but let's talk business for a moment). x86 migration is about opportunity. Even if IBM has initial success with the 970--there's no market for it. We've already been down this road before with the AIM alliance. Remember the "Burn baby burn" ads? I still have promo stickers that say "Snail Inside". (I can put those on my eMac now, BTW).
Supporting the release of the 970 as Apple's new "savior" chip is going to repeat a mistake that Apple has already made. IBM doesn't have the financial resources to compete with Intel. Regardless of "where" Intel and IBM say they will be at the same time technology-wise in the next few years, Intel will always come out ahead. They have the market. If Apple locks itself into PowerPC again, OS X, which Steve Jobs has slated as the "operating system for the next 15 years" is going to suffer. And in case anyone hasn't noticed, we've already been suffering.
From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs. They need a competitive OS to run these CPUs. And that OS is our OS X.</strong><hr></blockquote>
<strong>I've been reading this enormous thread and I have yet to see anyone approach this situation from a total business perspective.
The whole point about Apple and x86 isn't always about performance or technical considerations. (Yes, they are important, but let's talk business for a moment). x86 migration is about opportunity. Even if IBM has initial success with the 970--there's no market for it. We've already been down this road before with the AIM alliance. Remember the "Burn baby burn" ads? I still have promo stickers that say "Snail Inside". (I can put those on my eMac now, BTW).
Supporting the release of the 970 as Apple's new "savior" chip is going to repeat a mistake that Apple has already made. IBM doesn't have the financial resources to compete with Intel. Regardless of "where" Intel and IBM say they will be at the same time technology-wise in the next few years, Intel will always come out ahead. They have the market. If Apple locks itself into PowerPC again, OS X, which Steve Jobs has slated as the "operating system for the next 15 years" is going to suffer. And in case anyone hasn't noticed, we've already been suffering.
From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs. They need a competitive OS to run these CPUs. And that OS is our OS X.</strong><hr></blockquote>
Sorry my first post. Although it may sound reasonable to switch to intel or AMD for many reasons I dont too many Mac users will support it. I'd be willing to bet that less than 50% will. As illogical as that may be it will be the biggest gamble in Apple's history. Anyone want to vote on this?
IBM is almost four times bigger than Intel.</strong><hr></blockquote>
Quite true. IBM also makes some very good chips, has good manufacturing, R&D, and has a built in customer for their chips. They offer network solution in the corperate market where they can sell linex running on their mother board designs. They have a potential customer in Apple as well, and if they update their PowerPC-Open-Platform to include the new chips, and can interest a few third party linex boxe suppliers in these motherboards, then they might have more customers. Yes there are a few if's there, but the point is that if IBM wants to open the market up for the Power PC platform they can put quite a few resources behind the project. It will also be a bit easier than it was 2-3 years ago, since the open-source community and linex community are better established, and there seams to be a dislike for Microsofts licensing practices right now that if properly capitalized on by Apple and IBM could help as well. I think that the main quesiton is how expensive the computers from Apple and IBM will be for consumers and buisnesses when they are released. The way that IBM anounced the 970 poits to a direct push back into the desktop arena for the PowerPC platform backed up by IBM, and their investment in Linex, and (dreaming) possibly a closer alliance with Apple.
One point, when you are the leader in a market, it is easy to loose market share. Just look at palm.
Now that you've explained the prerequisite for any company wishing to compete in a marketplace, how does this explain IBMs possible strategy for competing with Intel?
<strong>Now that you've explained the prerequisite for any company wishing to compete in a marketplace, how does this explain IBMs possible strategy for competing with Intel?</strong><hr></blockquote>
To what extent does it make sense for IBM to compete with Intel? IBM is so vast that they have had identically named divisions selling different solutions to the same market, none of which were aware of the others' existence. They make several complete platforms, some of which compete with platforms that are powered by Intel CPUs, and some of which are powered by Intel CPUs. Most of the market they serve is heterogenous with respect to CPUs: There are (among others) Alphas, SPARCs, MIPS, the odd HP-PA, and Itaniums competing with IBM's POWER and PPC lines. Further down, IBM will cheerfully sell you workgroup servers and PCs running on Intel chips, because in this space they're (sort of) competing with HPaq and Dell, but not really - these machines are as much there to offer customers an all-IBM solution from desktops to mainframes, with whole huge suites of software thrown in for good measure. IBM also competes with Moto in the embedded space, and of course for Apple sales as well. This is the context in which IBM addresses Intel.
In other words, JCG (and others) are describing business models because those are what matter in the end. Your "industry standard platform" is an illusion, and the only chip companies competing with Intel are the ones offering compatible CPUs. Microsoft has settled on an x86-based platform. Linux and BSD run on just about anything. Apple has consolidated on PPCs. After a disastrous flirting with x86, SGI has run back to MIPS. IBM runs on its own chips. All of these work to varying degrees because what matters in the end is whether the CPU fits the needs of the platform running atop it. Needs refer to both present needs, anticipated needs, and legacy needs (which is, I believe, a big reason that Windows has stuck with x86, and also a significant reason for Apple to stick with PPC).
Intel have not proved invincible, especially when their weaknesses are attacked. Already the news from NASA is that big, hot super-CPUs are undesirable for high-performance computing. Consumers are lackadaisical about upgrading, because they don't see what an otherwise identical 3GHz PC gets them over their current 2GHz PC, except for even more obnoxiously fast scrolling. Intel has hit snags before, and AIM has offered faster processors than Intel has before, but that's not even half of the whole story. You can't consider a CPU without considering the whole platform. And you have to consider that there are a whole raftload of advantages to a platform aside from the speed with which it crunches integers. In fact, this raftload is exactly what Apple has dedicated itself to since its inception. The occasional speed lead is certainly welcome, but it's never been necessary to the success of the platform.
As far as I can tell, there are no other advantages to adopting the x86 ISA. It's ugly and unwieldy, the vector unit is mediocre, the sheer heat of the processors severely limits design options (remember that industrial design is concerned with usability and ergonomics more than with looks, so this hurts), the processors are more expensive, Apple would have to deal with the full complexity of a commodity motherboard architecture, which would raise their R&D costs precipitously, slow them down significantly, and negatively impact the platform's stability, predictability, and reliability. Apple would be a small fry customer locked into adapting Microsoft's design choices to their own uses, Classic would break, and above all, the consumer would have this choice: All else being equal, do I buy the operating system that looks cool but which I'm not familiar with, and which isn't the one that everyone I know uses, or do I buy the operating system that I know, and that I know will be compatible with work, and with all my friends' machines?
Comments
The whole point about Apple and x86 isn't always about performance or technical considerations. (Yes, they are important, but let's talk business for a moment). x86 migration is about opportunity. Even if IBM has initial success with the 970--there's no market for it. We've already been down this road before with the AIM alliance. Remember the "Burn baby burn" ads? I still have promo stickers that say "Snail Inside". (I can put those on my eMac now, BTW).
Supporting the release of the 970 as Apple's new "savior" chip is going to repeat a mistake that Apple has already made. IBM doesn't have the financial resources to compete with Intel. Regardless of "where" Intel and IBM say they will be at the same time technology-wise in the next few years, Intel will always come out ahead. They have the market. If Apple locks itself into PowerPC again, OS X, which Steve Jobs has slated as the "operating system for the next 15 years" is going to suffer. And in case anyone hasn't noticed, we've already been suffering.
From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs. They need a competitive OS to run these CPUs. And that OS is our OS X.
<strong>I've been reading this enormous thread and I have yet to see anyone approach this situation from a total business perspective.</strong><hr></blockquote>
That's because it's a thread about the 970, not a thread about the future of the entire platform. Although it is starting to fray now.
[quote]<strong>The whole point about Apple and x86 isn't always about performance or technical considerations.</strong><hr></blockquote>
No, it isn't. And when this discussion comes up, as it has in threads dedicated to the subject, many other factors are considered, including cost, developer support and politics. And the general consensus seems to be that a wholesale switch to x86 is a bad choice all the way around. Opterons might pop up in Apple server solutions, but that would be the likely extent of it.
I won't rehash the reasons in this thread, because this thread is about the 970. Unfortunately, UBB's search function doesn't work.
If you decide to start another one, it would go in General Discussion, because the choice impacts not only hardware, but software, corporate politics and strategy.
<strong>I've been reading this enormous thread and I have yet to see anyone approach this situation from a total business perspective.
The whole point about Apple and x86 isn't always about performance or technical considerations. (Yes, they are important, but let's talk business for a moment). x86 migration is about opportunity. Even if IBM has initial success with the 970--there's no market for it. We've already been down this road before with the AIM alliance. Remember the "Burn baby burn" ads? I still have promo stickers that say "Snail Inside". (I can put those on my eMac now, BTW).
Supporting the release of the 970 as Apple's new "savior" chip is going to repeat a mistake that Apple has already made. IBM doesn't have the financial resources to compete with Intel. Regardless of "where" Intel and IBM say they will be at the same time technology-wise in the next few years, Intel will always come out ahead. They have the market. If Apple locks itself into PowerPC again, OS X, which Steve Jobs has slated as the "operating system for the next 15 years" is going to suffer. And in case anyone hasn't noticed, we've already been suffering.
From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs. They need a competitive OS to run these CPUs. And that OS is our OS X.</strong><hr></blockquote>
MacLuv,
Good points. However, I keep coming back to OS/2. Back in the days of windows 3.X, OS/2 was a far superior OS than windows, it ran on Intel chips, it was backed by a much larger company (IBM vs Microsoft), and it ran windows apps with little or no problems. In addition, when OS/2 was competing with windows 3.x, it had no where near the market share it has now. And, I read somewhere that IBM spent almost 4 Billion on OS/2. So, I just can?t see how putting an Intel chip inside a fancy Apple computer case running OS X is going to be any different than what IBM tried to do. OS X compared to Windows XP is not as big an improvement as was OS/2 to windows 3.x was. And keep in mind that Apple makes most of their profit on hardware. How are they going to justify selling an Apple X86 box for 20-40% more than a comparably configured Windows XP box?
"From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs. They need a competitive OS to run these CPUs. And that OS is our OS X."
-----------
Maybe a few years into the future this could turn up as a real possibility for Apple. For right now, though, apple can not afford another huge change. OS X was a big enough change that for apple to make another one so soon would cause wide spread panic among the developers. Etc., etc., etc. I'm afraid that there will not Be x86 for apple in the near future and for that matter I suspect never. X86 is old and befuddled and not something apple will base their future on.
x86 migration is about opportunity. Even if IBM has initial success with the 970--there's no market for it. We've already been down this road before with the AIM alliance. Remember the "Burn baby burn" ads? I still have promo stickers that say "Snail Inside". (I can put those on my eMac now, BTW).</strong>
If Apple moves to x86, they will die. They will not be able to compete on price with Dell and maintain a competitive operating system on their own. If Microsoft didn't have an office app monopoly, they may have chance. But the reality is that MS leverages its office monopoly and operating system monopoly far too well. Maybe if OpenOffice makes a dent, Apple can transition, but its death right now. They will not be able to compete on price for something that is not different from other box makers.
<strong>IBM doesn't have the financial resources to compete with Intel. Regardless of "where" Intel and IBM say they will be at the same time technology-wise in the next few years, Intel will always come out ahead.</strong>
Yet, AMD, a smaller company then Apple, will be able to compete with Intel?
<strong>The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs.</strong>
AMD hasn't been competing with Intel in the last 3 months, and that lag will only continue to grow in the future. AMD may be dead in year, let alone compete with Intel.
<strong>
. . . Even if IBM has initial success with the 970--there's no market for it. . .
. . . From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. . .
</strong><hr></blockquote>
It sounds like you think the x86 has some special marketing appeal to it, and consumers will reject the IBM 970 because it isn't an x86 type processor. Can you tell us why you think Joe or Jane consumer will change his or her mind about the Macintosh just because it is running an x86? Really what they care about is whether it will run their software, and an x86 Mac will not do that either. People will usually switch to the Mac when it gives them something they cannot get from their present Windows PC. There is nothing magic about the brand name of the processor inside, as long as the computer does what they want.
Those who do care about technical things like the processor are interested in performance first, not so much brands and types. Consider whether this kind of discussion would even be taking place if the 970 initial clock speed was 6 GHz. You would be saying, "bye, bye x86," like everybody else. So, it is really a matter of performance, and we should pick the technology that will likely win the race, not just be ahead for the first few laps.
<strong>If you decide to start another one, it would go in General Discussion, because the choice impacts not only hardware, but software, corporate politics and strategy.</strong><hr></blockquote>
Feel free to move this post where you feel it fits.
--------------
wow, a lot to answer... :eek:
[quote]sc_markt says:
<strong>And keep in mind that Apple makes most of their profit on hardware. How are they going to justify selling an Apple X86 box for 20-40% more than a comparably configured Windows XP box? </strong><hr></blockquote>
Yes, Apple's profit design right now is based on perceived value rather than actual value. It is dependant upon people who do not yet realize PCs have become a commodoty market. It leeches off of what I call the Loyal Apple User Base (LAUB), locking us into a proprietary system that only Apple can control. Like you say, this means we pay extra for hardware that is outperformed by the competition--just because it looks pretty.
As far as selling an x86 box for 20% more... are you saying that we're paying more for the G4 processor itself? We're really paying for Apple's perceived value/goodwill. There is a great misconception that an x86 in the Apple PC would lower value, when it could actually raise it. How? That magic word Apple loves: perceived value.
As far as Apple restructuring its profit design--Apple already knows PCs are a commodity market, that's why they are setting up Apple stores. They just don't want anyone else to know the PC market is a commodity market. Steve Jobs is the man behind the curtain--and I think he's getting too comfortable back there.
For example, I don't know how much longer Apple can try and convince the public the G4 is a better choice over x86 in the consumer market, no matter how many "Genius Bars" they set up. <img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" />
-----------------------
[quote]From Algol <strong>
For right now, though, apple can not afford another huge change. OS X was a big enough change that for apple to make another one so soon would cause wide spread panic among the developers. Etc., etc., etc. I'm afraid that there will not Be x86 for apple in the near future and for that matter I suspect never. X86 is old and befuddled and not something apple will base their future on.</strong><hr></blockquote>
This is another misconception I always see from the "developers" point of view--that somehow another transition from Apple would cause a worldwide panic and engineers everywhere would rise up in disgust and burn the house down.
Unlikely.
From a business point of view, what has to be done, has to be done...
"You can handle people more successfully by enlisting their feelings than by convincing their reason." -- Paul P. Parker
Even if the migration to x86 seems unlogical from a developers' perspective, it can still be done. Everybody has a price.
----------------------
[quote]From THT<strong>
If Apple moves to x86, they will die. They will not be able to compete on price with Dell and maintain a competitive operating system on their own. </strong><hr></blockquote>
This is just not true at all. In order to convince me otherwise, you'll have to go into greater detail of why you believe Apple will not be able to compete with Dell on price or maintain a "competitive" OS. This argument always implies that Apple has seen its own shadow and should hibernate for the next fifteen years.
[quote]<strong>
If Microsoft didn't have an office app monopoly, they may have chance. But the reality is that MS leverages its office monopoly and operating system monopoly far too well. Maybe if OpenOffice makes a dent, Apple can transition, but its death right now. They will not be able to compete on price for something that is not different from other box makers.</strong><hr></blockquote>
Here I'm not clear on what you're getting at. Are you suggesting that MS Office provides the only differenciation between Dell and Apple? What about the "digital hub" Steve talks about all the time? What does this stuff have to do with the CPU inside the box?
[quote]<strong>
Yet, AMD, a smaller company then Apple, will be able to compete with Intel?</strong><hr></blockquote>
Yes, in fact, AMD has been competing with Intel for quite some time now.
As long as AMD produces CPUs, it will always compete with Intel within an oligopoly.
[quote]<strong>
AMD hasn't been competing with Intel in the last 3 months, and that lag will only continue to grow in the future. AMD may be dead in year, let alone compete with Intel. </strong><hr></blockquote>
What else do you pull out of that hat of yours? Rabbits? I'm not sure where you got this information from, but it is erroneous. As I've stated above, AMD, Intel, and any other company that produces CPUs are part of an oligopoly. As long as they all make CPUs, they are always in competition. I haven't seen any facts that AMD is pulling out of the CPU manufacturing business any time soon.
---------------
[quote] from snoopy <strong>
It sounds like you think the x86 has some special marketing appeal to it, and consumers will reject the IBM 970 because it isn't an x86 type processor. Can you tell us why you think Joe or Jane consumer will change his or her mind about the Macintosh just because it is running an x86?
</strong><hr></blockquote>
Firstly, why does one assume that when the 970 comes to market, Intel won't have something to compete with it?
Secondly, right now, the x86 kicks the G4s ass. Do you think another "Burn baby burn" revival is in order here? How long will it be this time around before Intel takes the spotlight over IBM?
[quote]<strong>
Really what they care about is whether it will run their software, and an x86 Mac will not do that either. People will usually switch to the Mac when it gives them something they cannot get from their present Windows PC. There is nothing magic about the brand name of the processor inside, as long as the computer does what they want.
</strong><hr></blockquote>
If this is the case, why don't we just slap an AMD Athlon in the iMac and call it a day?
Consumers care about a lot of things when it comes to computers--not everyone is the same. Something you don't mention is how quickly people want to use their software, which is always an issue in a commodity market.
[quote]<strong>
So, it is really a matter of performance, and we should pick the technology that will likely win the race, not just be ahead for the first few laps.
</strong><hr></blockquote>
Apple took a risk with RISC and it didn't pay off. Now you're asking me to have faith and walk down the same alley where we all got our asses kicked.
Unlikely.
IBM is in no position, and never will be, to infiltrate the market created by Microsoft and Intel. Becuase AMD manufactures a CPU that can replace Intel's, it is in a better position to compete with Apple as a partner. It will position Apple's OS right next to Windows, rather than having it in a completely different mindspace. Of course, this appears to be Steve Jobs' strategy, but things must change if Apple is to truly survive the next fifteen years.
[ 12-03-2002: Message edited by: MacLuv ]</p>
<strong>
Apple took a risk with RISC and it didn't pay off. Now you're asking me to have faith and walk down the same alley where we all got our asses kicked.
Unlikely.
IBM is in no position, and never will be, to infiltrate the market created by Microsoft and Intel. Becuase AMD manufactures a CPU that can replace Intel's, it is in a better position to compete with Apple as a partner. It will position Apple's OS right next to Windows, rather than having it in a completely different mindspace. Of course, this appears to be Steve Jobs' strategy, but things must change if Apple is to truly survive the next fifteen years.</strong><hr></blockquote>
I am not quite sure I understand the point of this very lengthy post. Put yourself in the consumer's role: you need a computer. You want to do this and that, whatever that may be, and you want a system that is going to be viable, cost-effective and do what you want. You can buy a Windows PC, a Mac, or whatever. The Mac accomplishes all the same tasks a PC does and vice versa. It's just a Mac does it in a more streamline fashion. Period. Listening to music, burning CDs, importing DV, burning DVDs, email, internet, and on and on. You don't have to worry that XP sp1 changes your drivers and now the task you did before doesn't work quite right, or hangs, or whatever. Seriously, I struggled with Windows or years, and had to reformat and rebuild my Windows PCs so many times, I just gave up on it. It is not worth my time and effort to go through all the nonsense with any Windows PC.
Thus we have the switch campaign. A brilliant move. They are not touting RISC is better in any advertising I've seen. And I paid around $2100 for my Powerbook. There's not a Windows laptop around that offers me what it does. I think it's ambiguous to argue what processor is inside a Mac or a Dell, or any other PC. It's what you can do with the computer that should make a difference, but people are so out of touch with how to really use a PC that is boggles my mind. The general public, and I hear it in conversations everyday, that they can get a P4 3.0GHz processor and blah blah blah. Who cares?! AMD, Intel, Motorola, IBM, I could give a shit, as long as I can work. And I can do that with a Mac.
BTW, have you used a new dual PowerMac? It screams! More a function of video technology that CPU at this point anyway.
<strong>
Apple took a risk with RISC and it didn't pay off. Now you're asking me to have faith and walk down the same alley where we all got our asses kicked.
Unlikely.
</strong><hr></blockquote>
I can see that you truly appreciate and respect the performance that Intel and AMD have managed to get out of the x86 type processor, but your strong feelings may be affecting your willingness to use clear logic. I too am impressed with what they have done with the x86, and it is truly an outstanding technical accomplishment. I think it caught everyone by surprise. And I think, along with others, it may have begun with the performance wars, which Apple started. I can see the head honcho at Intel saying, "Get us the highest possible performance and MHz, and I don't care how much it costs." Now everyone is playing catch up. In the end, this challenge will benefit us computer users greatly.
You say Apple took a risk with RISC (good pun too) and it didn't pay off. Think about that. Is RISC the fault, or is it the company who was making processors for Apple? Really. What would have happened if it was the other way around? What if Motorola was making x86 chips for Apple and Intel had the PPC architecture? Do you think for a moment it means Apple would now have a 3 GHz chip and Intel would be struggling at 1.25 GHz? You know the outcome of that scenario. Intel would still be beating Apple in performance, but by a larger margin in my opinion.
So, nobody is saying 'walk down the same alley,' which would be sticking with Motorola. IBM will do much more with the PPC than Motorola ever could do. I'm not criticizing the talent at Motorola, as I am criticizing the commitment of management to the workstation and desktop market, or even to the semiconductor business itself. There is nothing magic about the x86 architecture, and as others have pointed out, it has a lot of shortcomings. There is a whole lot more potential in the PPC path that Apple and IBM are on, IMO.
[ 12-03-2002: Message edited by: snoopy ]</p>
[quote]Firstly, why does one assume that when the 970 comes to market, Intel won't have something to compete with it?<hr></blockquote>
The PowerPC, whether from IBM or Motorola or Framistan Inc., is not, in any form, on Intel's radar. AMD is, period.
The G4 may be getting old and it's potential never came to its fullest bloom (yes, blame Motorola for quite a bit of that, but also blame IBM for being block-headed about Altivec). However, the G4 was a "good enough" step up from the G3 for Apple to keep its head above water and build OS X and all those iApps. Apple isn't a Dell (And me having to point that out is ridiculous). Unlike Dell or any other box assembler, the UI & software are closely tied to the hardware and vice versa. Do you think Shake was purchased with only the G4 in mind? I don't believe so.
And lastly, don't assume Intel is pulling its punches. They decided to win the war (over AMD) with gigahertz first, performance second and sacrifice power consumption and heat. This may come back to haunt Intel in the next few years.
But again, what Intel does is not relevant to Apple.
This is about what Apple is planning.
Screed
[ 12-03-2002: Message edited by: sCreeD ]</p>
Intel's own roadmaps says 3,4Ghz - 3,6GHz by then.
Will this be too much to beat for dual 970's? I doubt it, but single 970? It wouldn't surprise me if the 970 could teach the P4 a lesson in some cases
<strong>
From THT
If Apple moves to x86, they will die. They will not be able to compete on price with Dell and maintain a competitive operating system on their own. </strong><hr></blockquote>
[quote]Originally posted by MacLuv:
<strong>
This is just not true at all. In order to convince me otherwise, you'll have to go into greater detail of why you believe Apple will not be able to compete with Dell on price or maintain a "competitive" OS. This argument always implies that Apple has seen its own shadow and should hibernate for the next fifteen years. </strong><hr></blockquote>
This is amazingly simple.
Thought experiment: Assume Apple has successfully done all of the other steps they would need to do to make an x86 Mac. Answer a couple of simple questions:
Who is going to get a cheaper price on x86 chips? (Choose one: Apple, Dell) Note history of Intel disliking (for good reason) Apple.
Who is going to get a cheaper price on motherboard chipsets? (Choose one: Apple, Dell).
Repeat for all other components. The 86Mac would be cheaper (since all the components would now get mass produced at a larger scale), but they would NOT be cheaper _than_Dell_.
Dell has a 2% R&D burden, Apple has about an 8% R&D burden. On hardware. IGNORE the 'Dell doesn't make their OS' issue(s).
So Apple makes a sweet little $500 x86 Mac. Either (a) Dell can do the _exact_ same box for less, or (b) Apple puts flourishes into the design -> "perceived value" higher than "actual value" (your words).
_IF_ Apple switched, and I were Dell, I would want them DEAD. Wanting them dead is a sound business decision - a couple of years of mild hardship is a very small price to pay for a monopoly. I'll grant you that the $500 x86-Mac might be a nice box. But if I were Dell, I would sell a box no more than a couple of months after the intro of that 86Mac with precisely the same hardware - for at least $100 less. For _every_ product that Apple made that Dell could compete head-to-head on.
Apple has to either profit on the sales of these 86Macs, or gain lots of marketshare. Dell can cut the margin to 0 (there are lots of other product lines to pick up the slack). Dell can sell at a slight loss even - plenty of money stored up, and a clear advantage to killing off a competitor like Apple.
So assuming all of the other issues in the conversion to x86 cost $0, and 0 time... Apple would be making a box that is, you guessed it, costing more than the Dell. For people that enjoy using MacOSX, the extra price might be worth it, but it won't have the 'other 95%' saying "Wow, 86Macs! $500!" There wouldn't be any programs (of _any_ sort) that run faster on a 86Mac than on a Dell. When you eliminate product differentiation to compete in a DIRECT head to head competition you had better be selling more product and have a lower overhead. Neither of those will be the case for Apple in the next two years. Period.
That isn't where we were at the start of this year. There do exist real programs that have historically run faster on Macs - but those programs rely on the ppc or altivec! There are a couple of places full of dedicated PC lovers who repeatedly denounce the speed of the mac, and accumulate benchmarks to "prove" it. Fine. But some of those same sites _STILL_ have the top of the line Macs beating dual top-of-the-line Athalons in _some_ aspects of the benchmarks! Admittedly, it is in a short (and shortening) list of benchmarks that this is true, but there are places where a Mac is (or at least has recently been) the fastest machine for the job. But the 970 will at least reclaim dominance _in_the_places_Apple_needs. It might not be an overall faster processor when debuted, but things like photoshop benchmark tables are going to be re-written(even when run by PC guys). Just because the ppc970 doesn't (or won't) dominate _all_ benchmarks doesn't mean "That's a useless dead-end chip for Apple to adopt".
What I'm saying is, if I as a desktop computer purchaser needed a boatload of double precision floating point _today_, a dual athalon is the tool of choice. If I need parallelizable single precision floating point, the G4 is still up there. There are clear, objective reasons to purchase alternatives to Intel. If Apple did switch, we'd first be bombarded by a year of whining about 'Jeeze Apple's overpriced and gouging folk!', then another year of 'Why bother paying the Apple tax, this box is identical & XP is close enough'. -> spiral of doom.
Name a boxmaker or OS vending company that switched to x86 (LOTS of choices mind you) that can be considered a raging success story because of that switch.
This is just not true at all. In order to convince me otherwise, you'll have to go into greater detail of why you believe Apple will not be able to compete with Dell on price or maintain a "competitive" OS.</strong>
Dell is the largest and most agile PC manufacturer in the world, has the most inroads to consumer, business and enterprise markets, and has very good deals for all of its parts. It is well on its well to becoming a monopoly PC manufacturer. (Yes, I think HP will lose.) Now, Apple is essentially Gateway but for the fact that it has expend it's own resources for an "incompatible" operating system.
<strong>This argument always implies that Apple has seen its own shadow and should hibernate for the next fifteen years.</strong>
There may some truth in the statement. It also could be that Apple doesn't have enough business to truly innovate. It doesn't seem to have much foresight either.
<strong>Here I'm not clear on what you're getting at. Are you suggesting that MS Office provides the only differenciation between Dell and Apple? What about the "digital hub" Steve talks about all the time? What does this stuff have to do with the CPU inside the box?</strong>
MS Office is the gateway for any operating system that doesn't want to be a niche operating system. Without it, no operating system will make any inroads into the markets. Every single business uses it. Every single school uses it. The information in MS Office documents is the most valuable things we have. No one will migrate without being able to take their data with them.
The digital hub is just new marketing to sell computers. Dell can do it just as any other PC manufacturer can.
<strong>Yes, in fact, AMD has been competing with Intel for quite some time now.</strong>
The only win was the Athlon in 2001 with their 0.18 micron fab. Since Intel got to 0.13 micron first, AMD has been losing.
<strong>As long as AMD produces CPUs, it will always compete with Intel within an oligopoly.</strong>
When one manufacturer has 80% of the market, if not more, it isn't an oligopoly. Intel has all the advantages right now since they'll get to 0.09 micron and 0.065 micron first.
<strong>What else do you pull out of that hat of yours? Rabbits?</strong>
News articles:
<a href="http://www.forbes.com/newswire/2002/11/19/rtr799607.html"; target="_blank">AMD to move beyond PC, faster chips no longer key</a>
Reuters, 11.19.02, 4:04 PM ET
By Reed Stevenson
LAS VEGAS, Nov 19 (Reuters) - Advanced Micro Devices Inc. (nyse: AMD - news - people) said on Tuesday that it would embrace a strategy of developing processors for a wider range of products outside computers and called on the industry to focus on user needs rather than creating "technology for technology's sake."
...
AMD, which has fought a losing battle in recent quarters against Intel Corp. (nasdaq: INTC - news - people) for the top spot in processor speed and performance, said that making semiconductors smaller, cheaper and faster was no longer the key for an effective strategy.
...
AMD's strategy shift came on the same day that Standard & Poor's downgraded the chip maker on concerns about its ability to boost profits and weather soft demand, only hours after AMD said it would raise $300 million in a convertible note issuance.
<a href="http://www.theworkcircuit.com/news/OEG20021114S0026"; target="_blank">Troubled AMD cuts 2,000 jobs amid mounting loses</a>
By Mark LaPedus, Semiconductor Business News
November 15, 2002 (6:35 a.m. EST)
SUNNYVALE, Calif.--Reeling from the IC downturn and competitive pressures from Intel Corp., troubled Advanced Micro Devices Inc. today announced its much-anticipated move to reduce its headcount by 15%, or approximately 2,000 positions, by the end of the second quarter of 2003.
...
Hit hard by a slowdown in its core processor business, AMD recently reported sales of $508.2 million for the third quarter of 2002, down 34% from $765.9 million in the third quarter of 2001 and by 15% from $600.3 million in the previous period.
The Sunnyvale-based company also reported a net loss of $254.2 million, or minus $0.74 a share, compared to a loss of $97.4 million, or minus $0.28, a year ago. In the previous quarter, it reported a loss of $184.9 million, or minus $0.54 a share.
<a href="http://www.extremetech.com/article2/0,3973,563068,00.asp"; target="_blank">Update: AMD "Paper Launches" 2700+, 2800+ Athlon XP</a>
AMD "Sampling" 2700+ For PCs Available In November; 2800+ Only Available To "Enthusiast" OEMs
By Mark Hachman
While Advanced Micro Devices announced its 2700+ and 2800+ Athlon XP microprocessors today, customers may be asking when they can buy the chips the company announced in August.
AMD, Sunnyvale, Calif., will launch the 2800+ Athlon XP today in "limited quantities", a phrase that analysts say hearkens back to the race to 1-GHz?when AMD, not Intel Corp., was the company that broke the tape with volume quantities of 1-GHz chips.
...
McCarron said it's possible that AMD's development of the 0.13-micron silion-on-insulator process, combined with the larger cache sizes of the delayed "Barton" processor, as well as the need to develop working samples of the Opteron and 64-bit Athlon processors, may be hindering the company. "They have a lot of balls in the air," he said.
...
"Ultimately if they can't deliver it will show up in their revenue results," Brookwood said. "One has to assume there's a method to their madness. In the case of the 2800+ the goal is cachet, exclusivity. They're only there for serious gamers." AMD's "enthusiast" OEMs will include ABS, Alienware, Falcon NW, MicronPC and Voodoo PC.
The bottom line, analysts say, is that AMD is still wagering the company on Hammer?the latest Athlons are just a side bet. Still, AMD's newfound reputation for quality may be showing signs of wear, and the fickle hearts of enthusiasts may be shifting.
"They're clearly not sustaining the level of volume (at) introduction they did a year ago," McCarron said.
Lets see, the K7 is teetering out or their 0.13 micron fab has Moto-itis, they are bleeding money like no tomorrow with 6 consecutive quarters of losses with profitability predicted 9 months away, they are showing classic signs of a company in trouble with paper launches and "strategy diversification" in the face of competition and their savior product delayed.
AMD is not in good shape right now. And if they stay this small, they aren't going to get to 0.09 and 0.065 micron fabs quick enough to compete with Intel.
<strong>IBM doesn't have the financial resources to compete with Intel.</strong><hr></blockquote>
IBM is almost four times bigger than Intel.
<strong>I've been reading this enormous thread and I have yet to see anyone approach this situation from a total business perspective.
The whole point about Apple and x86 isn't always about performance or technical considerations. (Yes, they are important, but let's talk business for a moment). x86 migration is about opportunity. Even if IBM has initial success with the 970--there's no market for it. We've already been down this road before with the AIM alliance. Remember the "Burn baby burn" ads? I still have promo stickers that say "Snail Inside". (I can put those on my eMac now, BTW).
Supporting the release of the 970 as Apple's new "savior" chip is going to repeat a mistake that Apple has already made. IBM doesn't have the financial resources to compete with Intel. Regardless of "where" Intel and IBM say they will be at the same time technology-wise in the next few years, Intel will always come out ahead. They have the market. If Apple locks itself into PowerPC again, OS X, which Steve Jobs has slated as the "operating system for the next 15 years" is going to suffer. And in case anyone hasn't noticed, we've already been suffering.
From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs. They need a competitive OS to run these CPUs. And that OS is our OS X.</strong><hr></blockquote>
<strong>I've been reading this enormous thread and I have yet to see anyone approach this situation from a total business perspective.
The whole point about Apple and x86 isn't always about performance or technical considerations. (Yes, they are important, but let's talk business for a moment). x86 migration is about opportunity. Even if IBM has initial success with the 970--there's no market for it. We've already been down this road before with the AIM alliance. Remember the "Burn baby burn" ads? I still have promo stickers that say "Snail Inside". (I can put those on my eMac now, BTW).
Supporting the release of the 970 as Apple's new "savior" chip is going to repeat a mistake that Apple has already made. IBM doesn't have the financial resources to compete with Intel. Regardless of "where" Intel and IBM say they will be at the same time technology-wise in the next few years, Intel will always come out ahead. They have the market. If Apple locks itself into PowerPC again, OS X, which Steve Jobs has slated as the "operating system for the next 15 years" is going to suffer. And in case anyone hasn't noticed, we've already been suffering.
From a business perspective, it doesn't matter that the 970 may be a superior technology. It doesn't matter that it will support the path Apple's technology is locked into right now. Apple needs to get off this path if it's going to compete with Wintel. The whole concept of using AMD is that AMD is already competing with Intel. Even though AMD has told investors it's going to follow the technology and not the market, AMD will continue to make CPUs. They need a competitive OS to run these CPUs. And that OS is our OS X.</strong><hr></blockquote>
Sorry my first post. Although it may sound reasonable to switch to intel or AMD for many reasons I dont too many Mac users will support it. I'd be willing to bet that less than 50% will. As illogical as that may be it will be the biggest gamble in Apple's history. Anyone want to vote on this?
<strong>
IBM is almost four times bigger than Intel.</strong><hr></blockquote>
Quite true. IBM also makes some very good chips, has good manufacturing, R&D, and has a built in customer for their chips. They offer network solution in the corperate market where they can sell linex running on their mother board designs. They have a potential customer in Apple as well, and if they update their PowerPC-Open-Platform to include the new chips, and can interest a few third party linex boxe suppliers in these motherboards, then they might have more customers. Yes there are a few if's there, but the point is that if IBM wants to open the market up for the Power PC platform they can put quite a few resources behind the project. It will also be a bit easier than it was 2-3 years ago, since the open-source community and linex community are better established, and there seams to be a dislike for Microsofts licensing practices right now that if properly capitalized on by Apple and IBM could help as well. I think that the main quesiton is how expensive the computers from Apple and IBM will be for consumers and buisnesses when they are released. The way that IBM anounced the 970 poits to a direct push back into the desktop arena for the PowerPC platform backed up by IBM, and their investment in Linex, and (dreaming) possibly a closer alliance with Apple.
One point, when you are the leader in a market, it is easy to loose market share. Just look at palm.
<strong>
IBM is almost four times bigger than Intel.</strong><hr></blockquote>
So, how does this make IBM financially stronger to compete?
<strong>
{snipped for space, read original post}
</strong><hr></blockquote>
Now that you've explained the prerequisite for any company wishing to compete in a marketplace, how does this explain IBMs possible strategy for competing with Intel?
<img src="confused.gif" border="0">
<strong>Now that you've explained the prerequisite for any company wishing to compete in a marketplace, how does this explain IBMs possible strategy for competing with Intel?</strong><hr></blockquote>
To what extent does it make sense for IBM to compete with Intel? IBM is so vast that they have had identically named divisions selling different solutions to the same market, none of which were aware of the others' existence. They make several complete platforms, some of which compete with platforms that are powered by Intel CPUs, and some of which are powered by Intel CPUs. Most of the market they serve is heterogenous with respect to CPUs: There are (among others) Alphas, SPARCs, MIPS, the odd HP-PA, and Itaniums competing with IBM's POWER and PPC lines. Further down, IBM will cheerfully sell you workgroup servers and PCs running on Intel chips, because in this space they're (sort of) competing with HPaq and Dell, but not really - these machines are as much there to offer customers an all-IBM solution from desktops to mainframes, with whole huge suites of software thrown in for good measure. IBM also competes with Moto in the embedded space, and of course for Apple sales as well. This is the context in which IBM addresses Intel.
In other words, JCG (and others) are describing business models because those are what matter in the end. Your "industry standard platform" is an illusion, and the only chip companies competing with Intel are the ones offering compatible CPUs. Microsoft has settled on an x86-based platform. Linux and BSD run on just about anything. Apple has consolidated on PPCs. After a disastrous flirting with x86, SGI has run back to MIPS. IBM runs on its own chips. All of these work to varying degrees because what matters in the end is whether the CPU fits the needs of the platform running atop it. Needs refer to both present needs, anticipated needs, and legacy needs (which is, I believe, a big reason that Windows has stuck with x86, and also a significant reason for Apple to stick with PPC).
Intel have not proved invincible, especially when their weaknesses are attacked. Already the news from NASA is that big, hot super-CPUs are undesirable for high-performance computing. Consumers are lackadaisical about upgrading, because they don't see what an otherwise identical 3GHz PC gets them over their current 2GHz PC, except for even more obnoxiously fast scrolling. Intel has hit snags before, and AIM has offered faster processors than Intel has before, but that's not even half of the whole story. You can't consider a CPU without considering the whole platform. And you have to consider that there are a whole raftload of advantages to a platform aside from the speed with which it crunches integers. In fact, this raftload is exactly what Apple has dedicated itself to since its inception. The occasional speed lead is certainly welcome, but it's never been necessary to the success of the platform.
As far as I can tell, there are no other advantages to adopting the x86 ISA. It's ugly and unwieldy, the vector unit is mediocre, the sheer heat of the processors severely limits design options (remember that industrial design is concerned with usability and ergonomics more than with looks, so this hurts), the processors are more expensive, Apple would have to deal with the full complexity of a commodity motherboard architecture, which would raise their R&D costs precipitously, slow them down significantly, and negatively impact the platform's stability, predictability, and reliability. Apple would be a small fry customer locked into adapting Microsoft's design choices to their own uses, Classic would break, and above all, the consumer would have this choice: All else being equal, do I buy the operating system that looks cool but which I'm not familiar with, and which isn't the one that everyone I know uses, or do I buy the operating system that I know, and that I know will be compatible with work, and with all my friends' machines?
[ 12-03-2002: Message edited by: Amorph ]</p>