Spokeshave -

Obviously there isn't any formula but you don't need to be a rocket scientist to figure out that a working 52Meg SRAM is more difficult to produce than a 4Meg SRAM.

I believe that if you work the timeline backwards for Prescott you will conclude that Intel must have silicon now on 90nm process technology. As AMD has somewhat limited resources, compared to Intel, it would be hard pressed to be in development of a 90nm Hammer concurrently with an as yet non released 130nm Hammer and Athlon64. Money is tight and cash reserves are evaporating.

Maui -

I don't think so but I let my accountant handle those details.

EP

Keith -

I´d call it suicide.

For as long as I can remember AMD has been jumping out of the airplane while their parachute was still going through design signoff. Absolutely nothing would surprise me from AMD, except maybe profits.

Ten -

The 70-20-10 rule should apply here.

What's that?

Intel executive vice president Les Vadasz to retire; John Miner named president of Intel Capital


Published in M2 PressWIRE on Thursday, 17 April 2003 at 18:15 GDT
Copyright (C)2003 , M2 Communications Ltd.

SANTA CLARA, Calif. -- Intel Corporation today announced the retirement of Leslie L. Vadasz, executive vice president and president of Intel Capital, effective June 1.

The company also announced that John H.F. Miner, Intel vice president, has been named president of Intel Capital.

Vadasz, 66, was part of Intel's founding team in 1968. He held a variety of engineering and business management positions during his career at Intel. He managed the design teams that developed some of the semiconductor industry's most significant products, including the world's first merchant market large-scale integrated dynamic random access memory (DRAM); the first erasable, programmable read-only memory (EPROM); and the world's first microprocessor.

Most recently, Vadasz has been responsible for Intel Capital, Intel's strategic investment program, which he established in 1991. Vadasz was elected vice president of Intel in 1975 and served on Intel's board of directors from 1988 until 2002 when he reached the board's mandatory retirement age. He will remain director emeritus.

Vadasz has served on several government advisory boards that advise on technology issues, including the Presidential Advisory Committee for Information Technology and the National Research Council's Computer Science and Telecommunications Board. He also was elected a fellow of the Institute of Electrical and Electronic Engineers.

"Les Vadasz has been a source of innovation and strategic guidance throughout Intel's history," said Craig Barrett, Intel chief executive officer. "He is the ultimate engineer, with a knack for anticipating where the industry is headed and how to position us for success. Much of Intel's leadership position in the chip industry is the result of contributions Les has made."

Miner, 48, has served as Intel Capital's general manager since 2002, jointly managing with Vadasz the company's external investments, acquisitions and new business incubation efforts.

Prior to this, Miner managed Intel's communications products, server products, and desktop motherboard and PC building blocks businesses. Miner joined Intel in 1983, was named divisional vice president in 1995, and was elected Intel corporate vice president in 1996. He received a bachelor's degree in electrical engineering from Tulane University and an M.B.A. from the University of Oregon.

Intel Capital is one of the most active venture investors worldwide and has invested in more than a total of 1,000 information technology companies in more than 30 countries. In 2002, Intel Capital invested more than $200 million in more than 100 companies. Intel Capital focuses on making equity investments and acquisitions in support of Intel's strategic interests and invests in hardware, software and services companies in several market segments, including computing, networking, and wireless communications. Intel Capital currently has holdings in more than 475 companies.

Keith -

I have noticed the same "poet's license" used by YB but I can easily dismiss it. There is no malice there.

Joe -

I don't think anybody said that there will be 65nm chips sold at Newegg in 2004.

Not from AMD at least...

UnD

What is new here is that the direct comparisons to Intel frequencies is explicitly made.

You're right about this. In the past AMD has claimed the spec referred to how a device compared to an older model of Athlon. Nobody believed them but that was the claim. Now they've come clean, at least on what message they're trying to send.

YB -

Rolled those CCs to July. Made $0.90.

.9/7.5 = .12

1.12^4 = 1.5735 or 57.35% annualized return.

I'm pleased.... :)

SubZero -

Wednesday February 13,2002 12:36 am Eastern Time

AMD says on track to return to profit in Q2

Let's be fair, they didn't say what year did they?

YB -

Based on closing prices today, I can roll my April $7.5 CCs to July for a net gain of $0.75. That's an annualized return of 46%+. I have to keep in mind that I am operating this as a business and I'd be crazy to walk away from that return. If this same opportunity is there in the morning I'm going to roll.

EP

YB -

AMD will have 50 ingeneers at IBM site by the year end and they expect 55 nm production one year from now.

Isn't this the same company that expected to announce Hammer a year and a half ago? The same company that expected to be profitable this quarter? The same company that expected Hammer to be the world's fastest processor?

I expect AMD won't meet their expectations.

wbmw -

AMD might be waiting for the launch to disclose performance numbers.

Yes that's possible but nearly a year ago they were quick to release SPEC "estimates" for a 2.0GHz Opteron, which is still today unmanufacturable if expectations hold. We'll see next week if AMD's promise of the world's fastest processor holds. They promised better benchmarks than anything Intel has to offer. We'll see if that's true or not and we'll see if they can deliver real product.

EP

wbmw -

Many of the benchmarks have a grace period of several months before product availability where they allow an official submission to take place. Obviously, this means HP intends to ship such a system in the short term, July at the latest.

Yes, and for some strange reason Opteron has gotten certification yet...

EP

YB -

Elmer, stock is diving back to $7.5. So your problem is gone, you won't be assigned. Just don't write June calls right monday. They will cost much more by the end of next week. Hammer will move this stone to $7.8

Sounds like good advice. I hadn't considered the effect of the options opening Monday morning but you make a good case.

Thanks.

sgolds -

knew you couldn't resist answering! :)


What do you expect? You post a lie here about me and then accuse me of being the liar.

You're the liar, not me. You're on Dan3's level now as far as I'm concerned.

YB -

Did you write those CCs?

I'm walking a tightrope here. My April $7.5s expire this week and I don't want to get called away. I can roll them out a month and make $0.30 but I much prefer to go out 2 months but the Junes won't trade until Monday. With earnings (ha!) announcement we could take a nosedive back below $7.5 for tomorrows close. Tough decision here.

EP

Chipguy -

The Madison is a large chips (about 2x bigger than Opteron) but most of the area is L3 cache (~60% IIRC) and is protected by redundancy. And it is manufactured in a mature 130 nm bulk CMOS process so I have little doubt that Intel can turn these puppies out with good yield and in whatever quantities the market will absorb.

Good observation. And because so much area is sram the normal defect density predictions are likely to be too conservative. Intel will likely get higher yield than predicted by die size alone.

EP

fingolfen -

From the horses mouth:


http://www.hp.com/hpinfo/newsroom/press/2003/030416e.html

sgolds -

Elmer is just looking for any excuse to answer and repeat a lie

Up until now I had considered you a respectable poster but this kind of accusation isn't true.

Here's the article I referenced:

http://www.digitimes.com/NewsShow/Article.asp?datePublish=2003/04/15&pages=04&seq=19

Here is the quote in the article:

K.J. Chou, general manager of AMD Taiwan, said that samples of the Opteron processors have been sent to customers, and volume production will start shortly after the official launch.

Every time I refer to that quote I say "if true".

You say the my argument has been discredited.

It doesn't matter that his argument was soundly discredited and exposed

Where and by whom? We have an AMD executive quoted above and where has that been discredited? If you can disprove his statement then show us a credible source or clearly you are the one who is making false statements.

Haddock

Are you sure you are not speculating on the basis of rather little here?

Gee you're just too sharp. I can't get anything by you...

Dew -

unless there is a proxy battle, you cannot vote a director out of office. All you can do is withhold your vote, which makes no difference to the outcome.

I know.. I know. Well it was still fun anyway.

Semi -

So tell me, has ANYBODY been able to find an article where this "alleged" glitch has been uncovered by testing?

I'm searched and I can't find any reference.

wbmw -

No offense intended but only a fool argues with a fool...

Got my AMD Annual Report in the mail today, along with the shareholder voting proxy. It was a pleasure to vote to throw out the BODs and all the proposals.

Will they make the list of America's 10 worst boards again this year?

Economaniac -

re volume production. I don't think we can really read anything into these remarks.

I disagree. Assuming they are true, and that's unclear, it would indicate that AMD has yet to enter volume production. After 1.5 years of delays that would be horrendous. That means they don't yet know what problems they will incur in high volume. They don't know what problems they'll see with Athlon64 either. They just don't know. I can assure you, running small numbers of engineering wafers does not uncover all the issues that will crop up in volume production. We include "the phase of the moon" as one of the factors affecting yields, meaning the unknown factors that only show up over time and high volume. If (note this is an "if") AMD hasn't been able to start production as the AMD executive was quoted as saying, then this is terrible news and there's no way to put a good spin on it. It may not stop shipping small volumes of early production but it reinforces the fears that SOI was the biggest mistake in AMD's history. It has been discussed here that had AMD not chosen SOI, they should have been able to produce a bulk silicon hammer meeting the full frequency achieved by current shipping products. The expected fact that they will be only sampling small volumes of a significantly slower SOI hammer with reduced yields is a devastating indictment on SOI and is also exactly what just about everybody but "Where's mine?" Jerry predicted at the onset of the project.

Now let's see how this plays out. The quote of the AMD spokesman may be wrong. AMD may ship volume from the onset. All I've said here is based on a quote in an article that may not be true. We'll find out soon but my guess is that the quote was correct.

EP

Sgolds -

This is far different from Ephud's implication, which is that this is vaporware, which flies in the face of all evidence.

Excuse me but I think you should take issue with the AMD executive who said production would start after launch. Even a 6 year old could conclude that production didn't start before launch. That's vapor, pure and simple. I didn't make it up. I quoted an article quoting an AMD executive.

Dew -

I can't get through either.

Anyone listening to the CC?

Quarterly revenue flat year over year
Intel first-quarter revenue is $6.75 billion
Posted April 15, 2003


Intel has announced first-quarter 2003 revenue of $6.75 billion, down six percent sequentially and flat year-over-year.

First-quarter net income was $915 million, down 13 percent sequentially and down two percent year-over-year. Earnings per share were $0.14, down 13 percent sequentially and flat with the first quarter of 2002.

“Our financial performance for the quarter was solid with our computing-related business performing better than expected and our flash business coming in below expectations,” said Craig R. Barrett, Intel chief executive officer. “Two major announcements in the quarter highlight our continued, substantial investment in new products and technology, and our strategy of driving the convergence of communications and computing. We introduced Intel® Centrino™ mobile technology that brings integrated wireless capability and longer battery life to a new generation of mobile PCs. We also announced our innovative wireless-Internet-on-a-chip technology, code-named Manitoba, for cell phones. Both of these key products were well received in the marketplace.

“These types of leadership products, along with the scheduled ramp of our 90-nm process technology in the second half of the year, help to position us well for future growth.”

The fourth-quarter 2002 results included a tax benefit of approximately $75 million related to divestitures, which increased earnings per share by $0.01. Last year’s first-quarter results included a pretax charge of $155 million related to a litigation settlement agreement, which had an after-tax impact of $0.015 per share.

wbmw -

Summary of 3GHz Pentium 4 performance

SPEC. 1152 for SPECint and 1201 for SPECfp.

Awesome numbers! Congrats to all Intel Engineers associated with it. Now fix that anomaly guys!<G>

Ten, indications are that this anomaly affects a small percentage of units. If so, it should be possible to screen for it. A design fix would also be done as a long term binsplit enhancement.

EP

NAS -

If it isn't true then you are just stirring up FUD.

I didn't post that article. It was [supposedly] a direct quote from an AMD executive and I clearly posted the caveat "if this is true". I only pointed out the obvious implications. I'm sorry you don't like the implications but I didn't make them up.

If true, you might ask yourself why a product 1.5 years late isn't even in production at launch, and what that might mean. I could offer a few opinions but I would be speculating.

NAS

I find some of your comments to be false or misleading.

You are welcome to challenge me at any time you think I make a false statement.

Chipguy -

think the fact that this part doesn't set a new high for CPU frequency tends to argue against this cause. Intel has been shipping a slightly faster part (3.067 GHz) for quite a while now. I think it is more likely related to signal integrity or long term reliability

Entirely possible you're right. I'm only guessing but the bus interface to core has changed, could be something there or the cache. Either way I would think a screen could be in place soon while a stepping is done to address whatever sensitivity there is. I don't expect a long delay, BWDIK?

EP

YB -

I wasn't the one who posted the article here's the link:

http://www.digitimes.com/NewsShow/Article.asp?datePublish=2003/04/15&pages=04&seq=19

It says nothing about a translation.

You know by yourself that it doesn't take too many wafers to have "volume" Opteron.

No I don't know that. It may take a huge number of wafers. All we know is it hasn't happened yet after years of waiting.

Ten -

If someone repeatedly makes false statements and fails to make any effort to improve his credibility, I think it's fair to call him a liar.

I do too and I did, however you shouldn't call him a #%!@ liar because that's getting personal...

<G>

Paul

do you have any thoughts on what the most likely culprit could be for the recall?

Repeat after me....

There is no recall because product was never released..

There is no recall because product was never released..

There is no recall because product was never released..

Sgolds -

I didn't refer to the SARS scare or statements about news releases instead of an official launch. My post addressed the statement that volume production wouldn't begin until after the launch. Keith's dismissal, with no reference to support it, did not address the production question at all but for some reason you assumed it did.

No big deal, we all do it sometimes and sometimes it comes back to bite you.

Greg -

Good post, good info., except for the comment on the poster's veracity. We can figure that out for ourselves, without resorting to a personal attack. OK?

A person who knowing makes false or misleading statements is called a 'liar'. The word is clearly defined in the dictionary and is also a legal term used in our court system. That is a statement of fact and not a personal attack. If I added some flavor to the statement to enhance the term then that could be considered getting personal but a simple statement of fact should not be construed to be a personal attack.

EP