Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
I am simply providing an expert opinion on AMD's capability and performance
By your own admission, you have no information about AMD that is not public, nor do you have any "expert" qualifications to pronounce on matters other than your narrow technical specialty (if we believe even that). You are no more "expert" than many here who work in some capacity in the IT industry but have no special insight into AMD. The very fact that you tout yourself as an expert from behind an anomymous pseudonym is a very good reason to disregard everything you say.
AMD would have extreme problems producing it
There is not a shred of evidence to support this. Since release there has been a steady supply, a speed increase, and many reports of low-temperatures and overclockability.
Elmer, the share price has doubled this past year while hearing to you bite your nails about "manufacturability." Are you masochistically long in any other lousy companies with subpar products they can't manufacture? (Besides Intel, I mean.) I need a place to put all this cash ;)
No doubt that MS is on MS's side in the long run. I have no doubt either that Microsoft will eventually develop a Windows for as many instruction sets as Intel feels like producing (even Yamhill if that is released). That said, I think it is a little naive to think that MS does not try to shape the market in a way that suits their interests (rather than merely passively hoping it turns out a certain way as you seem to suggest in your post). My point is that their interests coincide with AMDs in this case, not that there's any bond of loyalty.
I submit that AMD64 helps them alot more than Itanium, and that MS would be much happier in a world where both Intel and AMD were producing AMD64 chips for the desktop and servers (assuming they remain competitive with IPF).
The best situation for AMD is the one you give, with Intel on IPF and AMD on AMD64.
I can't help thinking about this news that Microsoft won't support an alternative 64bit desktop instruction set from Intel. Otellini suggested 2004 for desktops, but MS doesn't seem to know a thing about it. Is there a game of chicken going on between MS and INTC regarding IPF on desktop? Or is INTC quietly planning to launch AMD64 processors?
In any case, it is clear that MS is on AMDs side (for now), and that could be enough. My guess is that MS is speeding up the WinXP launch to make sure the Athlon64 is well-established before Intel tries to put IPF on the desktop. It's in their interest to have a competitive market for CPUs based on open standards.
They need something that offers the high-performance to use all of the new features that they need to justify new products and upgrades. In other words, they need 64bit.
On the other hand, they want a backward compatible product to keep old customers happy. They want an architecture that isn't controlled by a single company to keep the prices low for a bigger market and to keep the innovations coming through competition. So they don't want desktop Itanium.
So when is the soonest that Intel could possible release an AMD64 cpu?
Doug, This is great news for AMD.
Agreed, best news all day by far.
Yeah, I remember that piece. I sure hope Intel doesn't adopt AMD64 anytime soon, because AMD will have to compete directly, albeit from a stronger position. The best scenario (for AMD) is if Intel sticks to Itanium.
INTEL´s president is endorsing 64bit for the desktop for next year. It will be interesting to see their 64bit offering for the mainstream - obviously next year.
So mainstream-64bit from Intel next year, which means IPF because there wouldn't be that level of software support for Yamhill that quickly, and I doubt they would adopt AMD64 unless really forced to. All that capacity Intel is building is to be able to make the big-cache dies that Itanium needs to be sellable.
AMD just has to be ready with a working process and manufacturing capacity to supply the sizable chunk of the market that Intel looks like it will alienate.
Thanks kpf--that makes sense. EOM
I'm not sure why many assume that Opteron and Athlon64 have the same die. Of course, they have the same core, but the die has to be different to enable non-registered memory. Otherwise, why else would the FX require registered memory? For that reason, it wouldn't make any sense to say that A64 is binning alot of sub-2ghz chips which are being sold as Opterons. Either it's not binning that low or AMD is rejecting the slow ones.
AMD expects to sell tens of thousands of the chips in the fourth quarter and hundreds of thousands in first quarter 2004. By mid-2004, it expects to be selling millions per quarter.
No source is cited for this info, and this is certainly not public info, and I doubt an AMD rep would disclose material info like that. Other rumors I've seen are c. 90,000 at launch, hundred thousands in Q4--i.e. same scenario, rolled back one quarter. The author didn't sound like he knew very much about the industry, but who knows...
Article underscores my concerns about manufacturing capacity; they need something more solid in this category, as much to reassure potential customers as to actually make chips.
Elmer, let me quote the lines from Keith's post that brought this whole issue up:
RUIZ: You asked me about any surprises and I told you a very positive one. The other one that was somewhat surprising to me on the negative side was the level of intimidation that our competitors have gone through trying to keep customers from really engaging with us.
Intel really deserves that all they´ve done over they years will heavily backfire on them one day. I sincerely wish them all the worst.
http://investorshub.com/boards/read_msg.asp?message_id=1463641
Confused? There isn't a thing in there that alleges anything illegal, though you might interepret it that way in your messianic fervor. "Intimidation" can cover alot of things that are perfectly legal. Moreover, Keith's assertion that Intel "deserves" payback does not imply illegality, just his annoyance at unspecified practices we have no reason to designate as illegal. Making your customers angry is not a crime, though it is a questionable business tactic.
Anyways, with respect to sgold's request, I'll drop this topic.
Elmer, Unless you convince yourself that Intel is doing something sinister to stifle AMD you would have to face the unfaceable. Despite all the technology gifts AMD has been one of the most incompetently managed companies in the history of the technology revolution. AMD fans will make any claim necessary to continue their denial.
This has always been the straw man of the Intel boosters here, but I don't think it represents anybody's actual opinion. If anyone thought that Intel had THAT much control over the marketplace, I doubt they would be bullish on AMD. Intel will do what it can to make things difficult for AMD, on that we can count--Intel shareholders have a right to expect nothing less. Personally, I think Intel has enough brains and lawyers not to cross the line legally (but go right up to it!). However, were I to find out otherwise, I can't say I would be too shocked. In any case, it is not a matter of blame or denial; it is the competitive environment in which AMD has chosen to play. I think everybody (except you) understands that.
Not really much else to say.
The legal term is libel.
It's only libel if it's not true. It has nothing to do with whether or not evidence is offered at the time of the allegation.
But, seriously, you're right. We shouldn't all rush to judgment based on countless first-hand reports and material evidence over many years now. I mean, look what happened with O.J. Simpson. We all thought he was guilty--and it turns out he didn't do it!
Things are looking better on the business side? That's news to me.
I'll reply as if this were a real question, and as if the recent rise weren't a sufficient answer. The whole year AMD has been around the 6-7 level with occassional devations. You have to remember that when Jerry bought last year, it was not at all certain to the general public that SOI actually worked, that AMD64 would be competitive (remember the 800mhz samples?) or manufacturable (and I *know* you remember that speculation). When I wrote this post http://investorshub.com/boards/read_msg.asp?message_id=1309848
that summed up my rationale for investing in AMD, the stock was still trading at pretty much the same level that it had when it looked like AMD couldn't manufacture an SOI paperweight, but by then (last month) it was clear that everything was on track and benchmarks were looking good. One of my rare good, timely calls ;)
Can you honestly tell me that AMDs situation doesn't look a whole lot better than it did a year ago, both in its technology and its financial situations going forward?
But our answers mean little. The big money has spoken already.
blauboad, why would you think short interest should be down now? I'd expect just the opposite. Short interest should decline in the early stages of a runup, but increase after its been running awhile. IMHO, this runup has gone on long enough without a correction. Thus I'd expect to see short interest peak about now.
Call it a personal prejudice; I've been stung bad by short sellers who knew things about the company that I didn't. But, generically, when short interest gets to a certain point, it makes buyers hesitate--a self-fulfilling prophecy. If these guys are all planning to exit at 10.5, it's no big deal, you're right.
I think short interest was one of the big factors keeping up in the 7s even as everything was looking better and better on the business side. It took alot of muscle to get us out of that.
Is that last month's figure, or do we know what the current (post-runup) short interest is?
I'd expect it to be down. Not too great if it is up.
I'll definitely agree that a deviant Bapco score alone means very little except to the really hardcore. Even on a somewhat geek-oriented site like fatwallet.com, I've seen that people had very little knowledge about what Centrino was or how fast it is relative to P4 or Athlon. One particular benchmark really doesn't matter all that much, particularly to the consumers who are buying these products. They go more by brand names and marketing. That's why Ruiz's new strategy is much better than old Sanders'.
However, AMD's problem is not "designing and manufacturing" CPUs. The designs are great, the manufacturing, in general, has gone well, too. I'll even say amazingly well, given the relative (to Intel) dearth of resources. It's more a matter of quantity and pricing power. OEMs will only design around a product for which they can be certain of having adequate and steady supplies (along with good marketing to boost demand). The marketing side looks very promising--benchmarks and HPC wins are really just marketing to the geek segment, after all. Now, if AMD ever manages to get their hands on enough manufacturing capacity to supply more than half the market or so, then they'll be unstoppable.
Given the lack of Intel response to AMD64, it looks like they are going with the Itanium-desktop strategy. The ONLY way this is going to work is there is no competitive x86 chip available in sufficient quantities. And VIA and Transmeta are not competitive (and not 64bit).
So the future success of AMD depends on
1) Intel sticking to Itanium strategy
2) AMD investing in decent marketing
3) AMD securing more capacity
4) AMD getting .09 right
Number 3 is what troubles me the most, especially after the fiasco with the foundry partner. The whole game now is about scale, and AMD needs to scale up or get rolled over. To address another argument that I've seen here, whatever AMD will end up paying for IBM to make its chips will be meaningless compared to the value that the increased production and IBM backing will create. I don't know whether or not it will be enough, however.
I'm growing more and more convinced that Sun will offer an Opteron server since their deal to use Suse linux (which included some unspecified "technology exchanges"). Sure, they will have to face product differentiation problems, but it's getting to be pretty clear that the market will differentiate Sun right out of business if it doesn't want to produce top-performing 64bit machines. A Sun win would be great for AMD's prestige and corporate respectibility, regardless of how many they sell. It would cause many potential buyers and OEMs to look at AMD more seriously. I wouldn't say the same for, say, a Gateway win.
The Athlon64 release is close enough to smell (pleasantly).
AMD has announced a new pricing policy with Opteron, and I would expect that the A64s will follow a similar strategy. Alot of rumors lately suggest that the launch will go well (speeds, quantity, OEM partnerships). Intel has delayed its competing chips, giving AMD a nice window (problems on .09 it seems, just as a few of us predicted). SOI is starting to show benefits.
If we can just push Intel's .09 chips to Q01, then AMD will own Christmas. Is there anything you can do hurt yields over there, EPhud? ;)
AMD will need to overclock its advertising and marketing--historically weak areas, though better since Ruiz took over. I'm hoping they get this right, because they have a chance to really create demand here. The usual MO is sell to an already existing market (i.e. Intel's), so advertising was not as as important. Now they have a unique product; If they want premiums (or parity at least), they need to hype it. Centrino shows you that how much people are willing to pay for good marketing.
All this should lead to profits. And I don't think we need to wait that long for this stock to get up off the book-value Jerry-bought-here bond-conversion level it's been sleeping on for months now. The expectation of profits that comes from a successful A64 launch with a couple OEMS on board will be enough--IMO--to double this puppy. The AMD64 strategy changes the business model to make it less dependent on and vulnerable to Intel, all of which should lead to a little bit more respect on Wall Street once it becomes clear that they're going to pull it off.
This is pretty much the same as my original rationale for buying late in 2002, and if anything, it seems more plausible now than then (back in the days of 800mhz Opteron samples).
I think there will be a nice window to sell before AMD has a chance to really screw up .09 :)
I would be curious to know if indeed it is only 32 bit x86 Solaris or if it is a 64 bit version.
Funny, I was thinking something along these lines with the recent news about the Sun and SuSE deal. Suse just happens to be the only distro that currently supports Opteron... coincidence? Probably. But Sun is the most likely candidate for Opteron's next OEM win IMO.
Re: Windows64 vs. Windows 2000
You guys should note the fact that, unlike Windows 2000, Windows 64 will only be able to run on new systems. Motherboard chipsets, integrated peripherals, video cards will work fine. The possible problem areas are things like old printers and scanners. And most of these external peripherals conform to generic Windows specs nowadays.
Edgarcayce, I think you're over-reacting. That's not to say that you shouldn't sell--that's your decision. Ruiz has done the costcutting that needs to be done so that if AMD64 is a success, AMD will profit and live on. If AMD64 is a dud, it doesn't matter whether or not they make or miss forecasts this quarter or next; they're dead in the water even if they had managed to break even this quarter. Anyone who is in this stock for the medium or long term is counting on a successful launch of a competitive Athlon64 and the increased acceptance of Opteron. If those things happen, AMD will succeed and the stock will go up. If they don't...
This made my day.
This is two guys from Apple giving an interview to hype the new dual-G5 desktop. Watch them squirm:
DMN: Now, you're saying it's the first 64-bit desktop machine. But isn't there an Opteron dual-processor machine? It shipped on June 4th. BOXX Technologies shipped it. It has an Opteron 244 in it.
Rubinstein: Uh...
Akrout: It's not a desktop.
DMN: That's a desktop unit.
Akrout: It depends on what you call a desktop, now. These… From a full desktop per se, this is the first one. I don't know how you really distinguish the other one as a desktop.
DMN: Well, it's a dual processor desktop machine, just like that one.
Akrout: It's not 64, then.
DMN: Yes, it's a 64-bit machine with two Opteron chips in it. It started shipping June 4th.
Akrout: That we'll double check, but in my mind, it wasn't.
http://www.digitalvideoediting.com/2003/06_jun/features/cw_macg5_interview.htm
Jerry R -- It's interesting that the Inquirer worded their sub-headline like this. There are a lot of companies who would drool over having a gross margin of only 50.9%.
Bias in journalism...
The Inquirer has an ironic sense of humor--something that Brits tend to think Americans lack.
sgolds My conclusion is that there is no reason for AMD to use the 940 for Athlon64, unless they did something stupid in the 754 definition. I posted all the references.
One reason you don't consider: AMD might do something like this to satisfy the mobo makers. AMD64 market is still so young that 3rd party vendors have a reason to balk at further artificial segmentation, at least before they break even on their AMD64 products.
yb re:940 pin A64
Why do you think that a 940 pin A64 is any better of a gaming chip than a 754 pin A64? Do you think mobo availability or quality will be a problem for socket 754?
If it would be possible to run an 754pin A64 on an Opteron mobo using an adapter--why would AMD even bother with two different sockets? The processors are differentiated enough in RAM and HT link specs. Seems like one socket would make it easier for customers and 3rd party vendors. It sure seems better than having two (or three!) A64 socket types.
[EDIT: Thanks sgolds, I just read your earlier explanation. Seems a 754 to 940 socket converter would be unfeasable, though it really doesn't matter for my purposes.]
drjohn
Blauboad your arguments have no consistent logical thought
Process . . . Please don't take this as a personal attack
I don't have to, because you're only making yourself look bad.
The Thread subject and argument here was that the centrino margins where not so good.
I never commented on the margins of individual Centrino units.
In what you are quoting, I only talked about sell-through, and indirectly, yields.
What do you think the is likelihood of a relatively new product becoming stagnate,
It happens all the time. Dell's whole business model is built to eliminate this problem. That would be a funny thing to do if it weren't ever a problem, wouldn't it?
furthermore the price premium as you refer
to it appears to be holding as would be expected for the ramp perhaps even slightly better.
That is EXACTLY my point.
Why did you even bother to make such a statement in the first place,
Because it is true, and you don't contest that.
can you see how it may be interpreted by someone else that the person making the statement lacks an understanding in the basic dynamics of CPU pricing.
I can see how it might be misinterpreted that way. But that isn't my problem.
And by the way, I have seen AMD rebates on some Athlon model processors. I'm sure Intel has done it at one time or another too. Maybe they don't have your understanding of the "basic dynamics of CPU pricing"?
I won't be replying to you on this matter again.
Elmer-- Barton's problem
Then present a case and we can debate whether or not it is plausable.
A couple things suggest good Barton yields/splits.
Management has said as much in the conference calls.
The chips are extremely overclockable to the extent that all grades below 3000+ appeared to have been downbinned for demand reasons. The fastest chips are also good overclockers, which suggests that there is alot of headroom in the process.
AMD is about to release the AMD64, and I expect that they will want it to have a clear performance advantage over the fastest Bartons. That gives them a reason not to push Barton as far as it will go, since in September it will become a budget chip and its premiums evaporate. It would be a difficult marketing position to have your A64 3600+ priced several times higher from a concurrently selling Barton 3600+ that, based on the QS rating, should perform identically well. So, they may lose a few high-end Barton sales in the meantime, but AMD is betting the company on Hammer already anyway. What's a few more chips in the pot? Like I've said, if Barton does get released in a higher grade/s, that indicates very good things for A64.
No reported shortages or acquisition difficulties for Barton.
I have more confidence in my bias than yours
I couldn't agree more :)
drjohn--"Sorry your last post indicates you don't have even a basic Understanding for CPU/platform reseller dynamics."
My only point is that if Centrino were stagnant in the channel, the price would come down one way or another, and fast.
Intel does not give $200.00 rebates, they just cut prices and
still make money. If you read up on Moore's law you will understand how that is possible.
So you don't agree that it would be a bad sign if Intel were to offer a $200 rebate? Because that's all I said; I never said that it's their usual practice or anything. But then again, I don't have "even a basic Understanding [sic] for [sic] CPU/platform reseller dynamics."
Of course I know eventually the price of Centrino will come down. I'm talking about a scenario in which it quickly loses most of its premium over P4 or Athlon notebooks. This can and does happen in a number of ways, but anyway.
Once Intel or AMD release better solutions, obviously the price of the current Centrinos will come down. I can't believe we're even discussing this.
To assume that Intel and AMD price the same way is to assume their products are subject to the same capability/forces. They aren't IMHO. AMD has a few enthusiasts who will pay a premium price. At the same time AMD has real difficulty producing premium products so they need to control demand. Intel may not have the same restrictions.
Intel is also restrained by demand and supply, albeit to a lesser extent given their dominance and resourced. If they weren't, the Celeron wouldn't exist. Best Buy would be all .65 dual-core Itanium clusters.
It seems to me that AMD's production is going quite well, certainly on the bulk process. IMO, Barton's problem is a demand and not a supply problem. But this is an educated guess.
In reality, we have no clue what yields are on Barton or P-M and no easy way to guage demand. Interpretation on this matter tends just to reveal the interpreter's bias.
Elmer--"So can you explain to me what you think "cherry-picking" is and how is a part treated that isn't cherry-picked?"
I didn't invent the term, and I only used it in anticipating what detractors might say--based on what they have said. I enjoyed your explanation, and I agree that it is not an especially good analogy to semiconductors. But the point is, I never used it as a technical term, anymore than "churning out Opterons" would mean that AMD is fabricating them from milkfat in wooden buckets.
Ask any reseller what products gives them the
best margins any they will tell you top of the line.
That doesn't mean they sell more of them. In fact, it usually means the opposite. While most retailers would be glad to sell only Prescott systems when the chip is released, reality dictates that they will need to stock many more lower-priced models to meet demand.
Whether or not Centrino is selling well probably depends on what you mean by "well." If Intel keeps the price premium on it, then that probably means that sell-through is very good. (Though when AMD manages to keep a premium on its high-speed chips, that can only mean yields are bad :)
Start to worry about Centrino when you see $200 Intel rebates on it.
Elmer--I've been testing parts for over 20 years
Then that's probably why you don't understand that terminology :) You see, it's an agricultural metaphor. As you may know, fruits and vegetables are not fabricated, but are organically replicated and then harvested (or "picked") by machine (higher yield) or by hand (labor intensive--lower yield). Hand picking allows for more thorough and careful evaluation and quality control than does machine harvesting; therefore you obtain better performing produce, albeit at lower yields.
It's always the accusation (from you too, as I remember) that AMD "handpicks" or "cherrypicks" its higher speed processors, with the implication that it cannot produce these is any meaningful volume.
Buggi, re:A64 Launch
1. Athlon64 will launch in two main types of CPU: Clawhammer, with 1MB L2, and Paris, with 256KB L2. ... The Clawhammer model will have two types as well. One model will have single memory channel support, and one with dual memory channel support.
This suggests decent binsplits, as I don't think AMD will release anything with less performance than the fastest Bartons for reasons I've posted before. If a 256k cache chip can manage that, it must be clocking well. And if they have that many at launch, it would be difficult to say they were handpicked. We'll see.
Hmm. Based on this, the only difference between Opteron 1XX and A64 will be the use of unbuffered memory. How important is that for workstations and servers, anyway?
2. Athlon64 will launch with a socket compatible to Opteron, in order to take immediate advantage of mainboards already in the marketplace, such as the Asus SK8N, and a few others. HOWEVER, further iterations of the chip will have pin(s) added so you will not be able to use them in Opteron boards.
I sure hope they don't make a mess out of the various sockets by making A64s in three different pin configurations (to believe the rumors)--unless they can use a socket 754. Anyone know if there'd be a problem with plugging a socket 754 chip into a 940 board--are the pinouts and pin config compatible and would the motherboard allow it?
One good thing about AMD shafting the mobo makers by delaying A64 is that you can bet the chipsets are going to be more mature and better performing, especially since several are already appearing for Opteron.
wbmw: Microsoft is leagues ahead of Linux in consumer mindshare, and it would take many years to even put a dent in that. The 64-bit label will do little for Linux sales in the consumer market, so Microsoft has nothing to fear. Linux has barely any marketing presence, and besides, it's not about how many bits the OS has, but rather how many applications run under it. Microsoft has all the cards here, and no amount of 'Droid logic can change that.
It's not the consumer market they are worried about. For a large subset of business desktops (not to mention servers), Linux is quickly becomming a very viable--and relatively free--option. All typically used applications have their open-sourced equivalents. Europe is on this in a big way. Microsoft is not oblivious to this.
Sgolds has shown that Intel did promise high-end workstations based on IPF. Now what about mass-market desktops and laptops?
So far as I know, there is no direct confirmation by Intel of IPF for the masses. But it's obtuse to throw out an idea just because it hasn't yet appeared in an Intel press release (though some on the board would disagree, I know). Based on at least the following facts, there doesn't seem to be much room to believe anything EXCEPT that Intel's long term goal was to migrate Itanium for the mass market:
1) Repeated denials of Yamhill since the issue has been raised
2) Statements admitting that 64bit desktops would be needed in the future (2010--a long way off, but still a stated intention of making them)
3) IPF having been named IA-64 (implying it is/was the ONLY 64 bit instruction set Intel had planned a la AMD64)
It seems obvious that Intel wanted to establish Itanium in the server/workstation world and bring it (or something like it, i.e. not x86) gradually into the mass market--at least as its "PLAN A". The advantages for Intel are obvious, and it meshes very well with the kinds of decision Barrett has made while CEO. Very interested to see what will happen when he leaves. And no doubt there is a PLAN B, C, D, etc.
Speculation, yes, but not groundless. And anyone who plays individual stocks is by definition a speculator ;)
You simply don't get 64-bit computing for free, unlike what AMD's propaganda states. You have to spend the money to get the software infrastructure to have a true 64-bit solution.
First of all, the "64 for free" phrase which management has said a few times refers to the processor itself, which is priced as a high end 32 bit processor even though it can do more. Has nothing to do with software. Now, there is OS and software support available and in the pipeline, and it does not look as if there will be too much of a premium if any placed on the AMD64 ports, so yes, you do get a 64bit solution basically for free vs. a Xeon equivalent (please don't quote me prices for servers, that isn't the point).
As far as porting goes, you've hit upon one the biggest selling points of AMD64--the ease of porting to it from x86. It took one programmer one week to port all of the latest Unreal game. One guy, one week. Sure there's testing that needs to be done afterwards, but it just isn't this massive undertaking that you're trying to suggest. Makes great FUD, though.
You have to spend the money to get the software infrastructure to have a true 64-bit solution.
You're getting AMD64 and IPF confused.
Do you really believe there are a lot of customers saying they want 64-bit servers (or desktops, for that matter) but don't want to spend any money getting 64-bit software?
The point is not about "getting" software, it's about porting it; since there's little or no premium for AMD64 software, cost is essentially the same as for 32bit. And yes, I do believe there are alot of customers who would like to upgrade to 64bit in the easiest, most cost effective way possible. That's why I'm here.
Wasn't all this obvious when the decision was made to develop hammer in the first place? Add to that the extreme uncertainty around SOI and you have one of the most irresponsible decisions of the modern tech era. Once again the shareholders foot the bill for Jerry's vendetta.
While I accept that there are risks, I actually prefer the third party approach, and I think Ruiz is better at schmoozing with these companies than Jerry was. I think the risks are less than an in-house approach would be, and it will enrich the platform as a whole. It's a good thing to have alot of chipsets running around, more competition, more choice, cheaper overall systems. If there are enough partners it spreads the risk of a buggy product (excepting the CPU itself, of course).
The all in-house approach that Intel is going after with Centrino and Itanium is bad for consumers and alienates the very same companies that AMD is schmoozing. Now, companies go where the money is and even if Craig Barrett said nasty things about every CEO in Tawain, they would still make motherboards for him. The thing is, though, that if Intel crowds these guys out of the choicest markets, they instantly become AMD's best friends.
That's why I don't think it matters whether Intel is strongarming or not. If there's money to be made in Opteron, someone will step in. If Intel sees to it that there's little money to be made in Itanium/Centrino/P5,6,7, then they will go elsewhere. No doubt new companies will rise on the back of Opteron--like the Rioworks outfit Dan3 mentioned. Probably some others will play it wrong and fall off. The beauty is, AMD just stands back and watches while these guys develop the best products they can. If AMD holds up its end and delivers quality products on time and in quantity (and that's the real "if" for me), then I don't really see the downside to having third party infrastructure developers.
Sure they do. And that's EXACTLY what the IntelEvangelists were saying a long time ago. Now, where were you a couple of months ago, to remind the AMD fans of that, when they were running around trying to convince themselves and everyone else, of intels "imminent" server demise, due to the Strong Launch of Opterons? I see that you were nowhere to be found. Now that their "story" is the exact opposite, I guess it's easy for you to criticize people who remind them of what they said then, isn't it? Welcome to the ranks of the AMDroid.
As a mindless AMDroid, I believe management unquestioningly when they tell me Opteron will not significantly add to the bottom line in 2003. I've never thought otherwise, nor do I remember any general sentiment on here or RB that Opteron would bring in mega-revenue before the end of the year.
I'm really glad you're here, though, reminding everyone of what you think they said and calling them names.
If AMD wants to launch a successful product, then it has the obligation to create all of the infrastructure necessary to make it a success. It was unwise to depend upon so many outside vendors to do so much of the heavy lifting for Opteron (i.e. chipsets, motherboards, compilers, software content, etc.) that it puts the probablity of success in jeopardy.
Without third party support AMD is dead. Designing its own chipsets, mobos, etc. would alienate the third parties that AMD depends on and divert much needed resources. This is one of those disadvantages that can turn out to be an advantage in the long run, as AMD does not have the option of making the mistake Intel is making by trying to become more proprietary and in-house. Of course, if key parties leave or produce bad products, the strategy could drive them into bankruptcy too. But they don't really have any other choice, do they?
I guess the question is did the Solectron cancellation actually result in less Opterons being sold last quarter. Or rather, how many less.
Anyway, it's peanuts at this stage. Everybody knows servers are a slow sell, especially on a brand new architecture, chipsets, etc. SemiconEng seems to think that the AMD brand name is so trusted and revered in the IT world that it could expect to skip a slow-ramp up, but alas, they have to get in line at the lab like everyone else.