超微半導體 (AMD) 2002 Q1 法說會逐字稿

Ladies and gentlemen, thank you for standing by and welcome to the AMD first quarter earnings release conference call.

During the presentation, all participant will be in a listen only mode. Afterwards, we will conduct a question and answer session at that time, if you have a question, please press the one followed by the four on your telephone.

As a remainder, this conference is being recorded on Wednesday, April 17 of 2002. I would now like to turn the conference over to Robert J. Rivet, Senior Vice President and Chief Financial Officer with AMD. Please go ahead sir.

Thank you

. Good afternoon and welcome to AMD's first quarter earnings teleconference call With me here today in

and hosting this call is Jerry W. Sanders, our Chairman and Chief Executive Officer. Also present are Hector de J. Ruiz, our President and Chief Operating Officer and Benjamin M. Anixter, Vice President, External Affairs.

I will begin by summarizing our first quarter highlights and then give turn it over to Hector and finally to Jerry. This conference call is a live broadcast and will replay via the Internet at www.streetevents.com and www.amd.com. The tape

replay number for North America is 800-633-8284, outside the US dial 858-182-6440. The code access to call is the same for both and that number is 204-54-132.

Before we begin the conference call, I would like to caution everyone that we will be making forward-looking statements of our managements goals, plans, and expectations. As you know, the semiconductor industry is generally volatile. Our product and technology development projects and our manufacturing processes are complex.

Current economic and industry conditions continue to make it especially difficult to forecast product demand because our actual results may differ materially from our plans and expectation today. I encourage you to review our filings with the security and exchange commission where we discuss in detail the risk factors in our business. You will detailed discussion in our most recent form 10-K filing with the SEC. Lets begin.

Our revenue for the first quarter was 902 million dollars, down slightly 5 percent from the fourth quarter of 2001 and down 24 percent from the first quarter of 2001. In our microprocessor product line, we sold 8 million units in the quarter and established a new record. We gained unit market share in quarter. Our microprocessor product line's revenues were 684 million in the quarter, down slightly 3 percent from the record level of the fourth quarter.

Overall, ASPs declined slightly to 86 dollars reflecting increased

success in the first quarter compared to the fourth quarter. Revenue in our memory product line was 160 million dollars in the quarter, declining 18 percent from the fourth quarter of 2001. Unit shipments increased 12 percent sequentially. We continued to reduce the breakeven point and on a 5 percent less revenues than that fourth quarter, I am encouraged to report

8 million dollar operating loss for the quarter.

Pro forma operating income was a loss of 8 million for the quarter, 10 million less than the fourth quarter, and 50 million less in revenue. Net income in accordance with generally accepted accounting principles for the quarter was a loss of 9 million or a 3 cents per share loss, an improvement of 2 cents per share from the fourth quarter. Gross margin on a pro forma basis was 36 percent for the quarter, up three percentage points from the fourth quarter of 2001.

Margin improvements were driven by continued cost reduction in all product lines, in particular qualified production in shipments of the 130 nanometer Athlon XP mobile products. Research and development spending was on plan at 172 million for the quarter and at record levels, up 7 percent sequentially. Marketing general with administrative spending was also on plan at 157 million for the quarter, down 4 percent sequentially.

Switching to the balance sheet, let us talk about cash, capital expenditures, and inventory. First cash. Our cash balance at the end of the quarter at 1.3 billion, up 400 million from the year-end, reflecting the 500 million dollar convertible debt offering we completed early in the quarter, and a positive 160 million cash flow from our operations. This was offset by new capital investments of 200 million and the purchase of Alchemy semiconductor for 50 million dollars.

We completed the quarter with 377 million of inventory, down 3 million from the fourth quarter of 2001. Inventories for both our microprocessor and Flash product lines are well positioned for the second quarter plan. Full year modeling

use the following; use 28 percent for the tax rate. Our capital spending plans for the year is the same as previously discussed and will be about 850 million dollars. Our investment plans include the completion of a 130-nanometer capacity in Fab 30.

We will convert all microprocessor production over to 130-nanometer technology before the end of the year. In addition, we will convert Fab 25 to Flash 130 nanometer capacity. Depreciation expense for the year will be about 700 million. R&D spending in the second quarter will remain at record levels. For the year, will use 700 million for our annual R&D investment. I will now like to turn this over to Hector to start the business segment discussion. Hector....

Thank you Bob.

Our first quarter was a very strong quarter operationally. Our unit shipments as well as manufacture units were at record levels for microprocessors and were sequentially up for memory product. Our productivity industries were a record numbers from our Fab throughout our assembly and test operations.

Our conversion rate to a 130 nanometer is ahead of plan and we now expect full conversion of Fab 30 to occur approximately a quarter ahead of schedule. Our

on a 130-nanometer SOI technology is progressing very well. Our

in the first quarter of our eight generation with Hammer processor has demonstrated fully operational hardware and we have begun sampling our key infrastructure and complemental partner.

The ramp, yield, and performance are in an excellent trend and our confidence level that we will shift Hammer for revenue before year-end is high. In co-ordination with our Flash partner Fujitsu, we continue our conversion of Fab 25 to Flash, which is progressing very well and we expect

to be 100 percent converted to Flash in the fourth quarter of this year.

We are very pleased with our MirrorBit Technology project and expect to be sampling customers this quarter and begin production shipments in the second half of the year. During the quarter, we established a partnership with

that will be

to our capacity needs as well as our technology development at the 65-nanometer nod and below as well as our high performance

process.

The unit at 130 nanometers, we expect to benefit from the foundry capability before the end of this year and in addition we expect to utilize over 300 mm foundry capability at 130 nanometers as well as 90 nanometers before our JV becomes operational in 2005. During the quarter, we acquired Alchemy Semiconductors, a provider of connectivity solutions based on

chip technology that has

architecture at its core.

Recently, we announced the

1100 product, which has the highest performance operates at 500 megahertz and lowest power operates at 250 milliwatts. PDA solution in the market. This

chip connectivity solution can support any of the operating systems used in this segment of the market. Product and technology leadership are keys to our success in our future. We continue to relentlessly invest in this arena and our success in 2001 with nearly 1100 awarded US

continues to validate our commitment and performance as a true innovator. I now would like to turn it over to Jerry.

Thanks Hector.

In the just completed quarter, the market acceptance of the Athlon XP processor continued to be gratifying. Remarkably, in the face of the most widely promoted product ramp in our competitor's history record unit sales of PC processors at 8 million units and gained about one percentage point in worldwide market share. Incidentally, during the quarter, AMD shipped its 200 million PC processors.

Last quarter, with some of industry shipments of PC processors at 41 million units, that is Microsoft Windows compatible processors from mobile desktop server and workstations computing and specifically excluding embedded applications such as Xbox. Some industry inventory replenishment was indicated. Going forward, such replenishment is unlikely to add to demand until PC end markets improve. No such improvement is yet evident. Accordingly, we expect that unit sales of PC processors will follow seasonal patterns this quarter, which generally show a sequential decline in the 5 to 10 percent range.

Under such a scenario, we expect to hold the market share at the current level. The current PC slump and the market share gains of Dell Computer

the exclusive province of our competitor, make further market share gains for AMD problematic in the near term. We believe our product line up for desktops coupled with the improving competitive position in mobile processors as a result of our accelerated transition to 130-nanometer production will enable us to grow sales significantly, when the end market for PC improves.

Noted that R&D spending is at record levels. The pay off is imminent. The single most important element in our strategy for growth independent of end demand is the Hammer family of processors. With Hammer, for the first time in our history, we will have a product line that covers the complete breadth of computing needs from the most powerful servers and workstations to desktops and mobile PCs. As Hector indicated, we have working silicon. Most notably, we have working silicon on the most complex member of the family internally code named Sledgehammer.

This devise is currently being produced on 130 nanometer silicon on insulator technology with nine layers of copper interconnects in our Dresden mega

. This 100 million transistor devise is targeted at the workstation and server market and will compete effectively against Itanium and high-end Xion processors. It should be on the market early next year. Clawhammer built on the same process and targeted initially for desktops, is expected to shift by the end of this year.

As Hector says, our confidence is high. Indeed, we expect to have initial samples of Hammer products on our next generation 90-nanometer technology node by the end of this year with production release mid next year, which will enable significant penetration of the mobile market. For the moment, our processor business is enduring a PC slump. Our overall business is unlikely to show substantial improvement until PC demand improves. On a brighter note and a clear reversal fortunes, our Flash business is looking much better.

We expect a substantial increase in revenues this quarter with momentum going forward augmented by our enhanced competitive position at the leading cell phone providers as a result of the MirrorBit Flash road map. Flash, which has been a drag on earnings of late, should be making a positive contribution going forward. With that, I will open it to

would you arrange the Q&A please.

Ladies and gentlemen, if you have a question, please press the one followed by the four on your telephone. You will hear a three-tone prompt to acknowledge your request. Your question has been answered and you will

your request you may do so by pressing the one followed by the three. If you are using a speakerphone, please pick up your handset before entering your request. One moment please for our first question.

The first question will come from Adam Parker with Sanford C. Bernstein. Please go ahead with your question.

Hi, You said that there was some substantial growth in the server market in the quarter. I was wondering if you could help me a little with your forecast for 2Q, maybe discuss higher sales and marketing strategy for servers is different from the desktop and mobile strategy? Then let us have a quick followup.

What I see in the server space is we're just getting started. We have only been shipping in the server market for a couple of quarters and so you know our unit volumes are in the 100,000 range. So, our market strategy is to find people who are looking for the highest performance, servers at the lowest possible cost.

We're building up our infrastructure support. We're forming our efforts primarily in the North American market for the customers who are most familiar with us. So, we're looking for the low hanging fruit first. So, I think that we continue to believe that by working with the so called white box manufacturers and getting a support base on our performance at a cost strategy, we'll be successful. But our real trust in the server space doesn't really come until we get to Hammer.

Okay.

Just switching gears a little bit, can you talk a little bit more about Flash in terms of some more color on your gaining share from the leaders there? Was it particularly strong in Europe or was there a geographical breakdown for the unit strength in Flash?

I think what we see there is that certainly last quarter, we did not see strength in Flash. Last quarter was the Z`

as far as we could tell. The reason we're optimistic for the current quarter, the reason we expect substantial sales growth is because we have made some market share gains in the designs with the leaders.

You know who the leaders are, they are Nokia, they are Motorola, they are Siemens, they are

Ericsson, and they are Samsung. We're not going to tell you exactly which ones we're gaining share with, other than say that it's among those five guys. I think the biggest contributor to our share gains is not just our current product line, but their confidence in going forward that AMD is going to be a long-term provider of the lowest cost per bit solution, which of course is what drives them as they put more and more and more per

into the cell phone market,

each individual handset.

I guess the only other thing I would say is that just across the boards; inventories are

worked down. So, we are seeing a recovery in the cell phone market. I think the rest of the recovery is primarily you know just people now having inventories in line with their usage and with an expansion in the economy, Flash as you pick with this product.

Right okay. If I can, just one last question. I asked Intel just yesterday, but I am wondering if can just tell me a little bit about the performance and compare your 4Q, which seasonally is the strong quarter in terms of how the 3400 Athlon Hammer would work vs. say P4 3

or whatever they are saying they are going to have out in 4Q?

Well, it's our intention to have competitive offerings throughout the lifetime of everybody at this table. So, hope that is a long time. Basically, our model numbers can be roughly equated to being superior in performance to an equivalent gigahertz rating.

So, if had a three point....excuse me if we were competing with 3G P4, then a 3400 plus model number, would be substantially better. We won't get to the 3400 model number without Hammer.

Right.

That is targeted at the end of 4Q or ............

That is targeted for first productive shipment to the end of 4Q, substantial revenue contribution in the next year.

Okay, thanks very much.

The next question will come from Scott Randall with Soundview Technologies. Please go ahead with your question.

Joe, I wonder if you can talk a little bit about how you're looking at trading off market share vs. sort of profitability, and specifically

down about 4 percent. You mentioned Dell might have

gaining share. What are you thoughts right now in terms of how much share are you trying to gain vs. either trading or pricing to do that?

Well, as I said, we don't expect we are going to be out. If we had a market that isn't growing or possibly even shrinking, that's an extremely tough market in which to gain market share. It looks like the guy who's gaining market share overall in the PC space is Dell, is currently in exclusive Intel house.

So, our expectation is that because we have an enhanced mobile position as a result of our 130 nanometer production and because we're substantially increasing the availability of that product in the current quarter, we think that any set backs we might have in the desktop, which is currently experiencing no growth, therefore, it is tough to gain market share, when the guy was gaining all the market shares.

Dell, we think that maintaining market share is a fairly impressive accomplishment in this mode. So, at the end of the day, we feel that we've been extremely successful in the developed countries. We like to stay the smarter you the better we look, but that will somehow sound

to something. So, what we would rather say is we just have to get the company familiar with our offerings. So, Intel is doing much better unfortunately than we are in China and India, where we don't have nearly the marketing presence they do.

Similarly, in Latin America, it is not where the......... they claim that their successes have been largely in those countries. Well those present new opportunities for us. What we found is when we show the customers what we have vs. what Intel has, we're successful. So, the fact that we've been able to increase our market share in the US and then hold our market share in Europe, I think it says a lot about our offering. So, right now, we're not proud of our achievements in China. We have to work a lot harder in China.

We really haven't put much effort into India and that is another opportunity. As far as trading off market share for profitability, we believe we've got an extremely good cost structure on our products. Our 130-nanometer devices are very small dye. As Hector has indicated, our operating efficiencies are excellent. We just need market opportunities. We're doing well where we have our marketing presence, where we are obviously the smaller company compared to Intel such as Latin America and China, we got some work to do.

Ok, thank you. Then two quick follow-ups. On the Flash side, significant unit and revenue growth. Do you assume that pricing is relatively defining some stability on 2Q?

Stability at

levels, but yes, stability.

Can you say what on the hammer

that you are sampling, what clock speeds are you currently sampling at?

Well, I don't think that we are going to talk about clock speeds here. I think the more important thing at this point, (clears his throat), excuse me, and sampling is do we have a functional part, so people could begin to build the infrastructure. You know, we're very pleased with the progress at having working silicon both in single and dual mode server motherboards.

The clock speeds are being limited at the current time by some circuit anomalies, which were debugging as we speak and we have high confidence that by the end of this quarter, we'll be able to confirm that we'll introduce

at least 2 GHz.

Great. Thanks very much.

Our next question will come from Evan

with Angelo

and Co. Please go ahead with your question.

Thanks. You said you guys lowered the breakeven point. Can you tell us, what that might be revenues in gross margin for breakeven?

Yea. This is Robert. As we talked about before with the 50 million drop in breakeven in sales from fourth quarter to first quarter, we did reduce a lot. We continued to make progress on that effort. We'll see more progress in the second quarter to sell around the 900 level or a little bit below as we make progress and we'll continue to make progress throughout the year as more things happen in the second quarter.

For example, in the second quarter, we will complete the closure of Fab 14, 15, which we announced in the third quarter of last year, which will help us reduce the breakeven point even further.

give you with specific granularity what the breakeven point is, but as you can tell from the current quarter, we were right at 900 million.

I understand.

And for Q2 guidance, it seems like the revenue down five to ten percent is mostly unit driven, typically seasonal. Do you have any type of ASP assumption? I am sure you are not in the habit of giving that, but any type of comment, is that down five to ten percent more unit driven or more ASP driven?

I think you know, that is the hard thing to predict.

Sure.

Because, as you probably are aware, it has already been publicized. Intel plans to be very aggressive in taking the price down on their Pentium 4.

Right.

Not knowing exactly what that is going to be and planning to, you know, certainly offer a competitive alternative, we are just reluctant to go anywhere with that until we know what the ultimate mix of our product is going to be.

Right.

Currently we're selling all of the Athlons that we are basically building. Certainly, all the high end Athlons and we had some success with Durons last quarter. But nearly 80% of our revenues came from Athlons. So, that encourages us that ASPs will hold up reasonably well. As we said in our press release, it's very unpredictable. My belief is that there was some inventory accumulation in the channels of Intel and AMD products and we're down to a two-horse race now, as you know.

Going forward, you know, the customers like this. They play

against the other. So, pricing is going to be under some pressure as long as business is as weak as it is. So, I think, we are just reluctant to say that we certainly don't expect ASPs to go up this quarter or we're not looking at any material changes.

Okay. Can you give us any indication of the market acceptance of the

and you guys have a nice partnership with Invidia. Can you give any commentary on that?

The world loves

.

Right.

This is a simply great product. You might have noticed that HP just announced a family of pavilion PCs using Enforce. This is a great product. We just need to get

out there. But, of course that you have also noted, the Enforce charges a little more money and right now, you have price sensitive market.

So, it's hard to predict exactly, you know, again, you should ask the people from

how are they going to trade off profit margins for market share. You

that a great product there.

Are you expecting any additional sales given the technical goodness of the product, where people are saying, I want to enforce chipsets so, there is

for a P4. So, you know, I am buying an AMD microprocessor then?

We certainly expect that to be the case and as I say, but you still have to have the customer walk into the store and want to buy something.

Sure.

We are more concerned about the number of customers who want to buy something than being able to persuade those customers to buy ours.

Thanks.

Our next question will come from Joe

with Merrill Lynch. Please go ahead with your question.

Hi. Guess that is Joe

recognize my name.

This is Jerry Sanders, Joe.

Yeah. Hi.Two questions.

First, Robert mentioned you are not doing a bit better given Intel's failure to really do anything with Celeron here and you are more competitive,

do you think this is perhaps where a lot of the market share gains are taking place? And then I have another question.

Well. As you noticed, we shipped more Durons this quarter than last quarter and frankly, we shipped slightly fewer Athlon this quarter versus fourth quarter. So, we had sort of a

distribution, Joe. The customer seemed to like our high performance

, we're selling all the model 2100 pluses and all that model 2000 pluses and now we are selling the lower end stuff and the Durons.

There doesn't seem to be any middle market. It's probably because the value seems to be bi-modal. You want performance, you get great performance using an AMD Athlon, and if you just want the low end, you know, you know you can get a Duron, which is better than a Celeron. So, going forward as Intel pushes their P4 into that Celeron stage, and presumably we'll figure out what to do with that value as brand Celeron. It will have an impact on what our mix is going to be.

Currently, I expect that our mix is going to be about the same. If anything, I expect, maybe we'll do a little bit better with Athlon this quarter and a little worse with Duron this quarter.

Okay. That makes sense. And then second question. Just to go back, you know, you guys are sort of

into Intel's product gap, very successfully, you know, what had been a year and half ago.

Do you have a gap here, Jerry, between the ramp of Hammer and what's currently going on, where you know, maybe Intel might be able, over the intermediate turn to take advantage of your width with the people in Northward

?

Well. I am less worried about the people in Northward

as I am about the fact that Intel has certainly been accelerating their rate of introductions of higher performance parts. To that end, Intel is certainly more competitive with us today than they were a year ago. I wouldn't deny that. I think, as I said, the

race, there will be times that we have an advantage, there will be times when they have an advantage. I think that there will be a toing and froing over the next six to nine months, but I think that the Hammer should once again give us a clear

in the PC desktop's phase.

Okay. Thank you very much, Jerry.

Our next question will come from Eric Rothdeutsch with Robertson Stephens. Please go ahead with your question.

Thanks.

Hi. Jerry. Can you give an idea what you saw for processor ASP trends during the quarter and if you could share what your blend of ASP was and then on the Flash side, can you give an idea what the Flash densities are of the design wins you have gotten at the cell phone manufacturers and what gives you the confidence that Flash will grow substantially in Q2?

Do you have backlog on the Flash side from the cell phone customers entering Q2?

Let me respond to the Flash thing first. Orders were stronger for Flash. Flash is very much a

business in much of our markets but it's on the strength of actual orders and backlog and customer indications that we're predicting a substantially stronger Flash quarter.

The multi design wins are 64 Megabit and 128 Megabit. We also have some design wins with some more complex devices going forward. As I said earlier, it's our roadmap, which is attractive. We've been able to consistently meet our roadmap. So, now that we've got 64 Megabit and 128 Megabit, and a roadmap that is competitive, we've been able to gain some share, we think.

We have also got this multi chip packaging capability for certain customers who want to put their

in with the high density Flash and there we've been, I think, quite successful. We see that you do very well with that. There really isn't a lot of, I can comment. On your processors question, would you repeat that one more time for me?

Yeah. I was curious what your blended processor ASP was for the quarter.

Oh, I am sorry. I failed to communicate that. Robert, what was it exactly?

86 dollars. 86 dollars, down from 90.

Okay. Great. Thank you guys.

Our next question will come from Hans Mosesmann with Prudential Securities. Please go ahead with your question.

Thank you.

Most of my questions have been answered, but I am curious about the Flash density expectations and wireless for the second half of the year and how you are positioned there? Thanks.

For competitive reasons, I don't think we want to comment there. We think we've got a competitive line up with Intel, who's the only real competitor and we think, going forward, we should have a lower cost per

than they do. At one time, Intel had virtually 100% of Nokia's business. Now we've got some and they're going to the 128 Megabit and we're going to participate in that business, we think, in a substantial way.

Would MirrorBit start representing a meaningful part of that Flash revenue?

Absolutely.

Okay. Thank you.

Our next question will come from Jack Geraghty with Gerard Klauer Mattison. Please go ahead with your question.

Jerry, just kind of a philosophical one related to what I said about the competitive landscape i.e. delegating market share. So, I guess,

if this is correct.

just waiting for the

to come back because, is the Hewlett Compaq thing causing sort of turmoil or is there some other thing in the marketplace besides just Dell doing something, that is sort of, I guess holding you back in a sense?

Well. It turns out. I don't think that anything is holding us back. I think the market is not growing and Dell is gaining share. That says the served available market of AMD without Dell is less. So, you could say in the non-Dell market, AMD is gaining share when we hold share. So, we're not waiting for the market to come back.

You know, we're accelerating our transition a 130 nanometer. We're working diligently on our hammer family to get early samples in the hands of infrastructure partners so that when the chips are available, we'll be able to ramp and not be limited by the availability of motherboards or chipsets, and of course, we have got the dual processor that is the Athlon MP, which has been able to make some in-roads into the server space and getting us to learning more about the server business.

So, all of those things encourage us to believe that we can gain share. But, with our market is not growing and Dell gaining the share in that market, I think we're doing remarkably well on the current conditions to maintain our worldwide market share. We have not given up on Dell. We think hammer is a great opportunity for us to penetrate Dell.

We also believe that the combination of Hewlett Packard and Compaq will be a stronger competitor for Dell and we think that we've got a good relationship with both those companies, which we hope would be enhanced by their merger, which we strongly support.

Fair enough. Just

one more. On an overall basis, would you expect somewhat of a `normal year` in the sense of your sort of having a little bit of a normal that you say is seasonal second quarter. Would you expect hopefully seasonal trends to exert themselves in the third and fourth quarters,

you can see right now?

Oh yes. I absolutely expect seasonal

back-to-school should be enhancements to the demand of course

the normal selling season at the end of the year. I, in a course, like everybody else who is in technology and IT, I hope that this replacement cycle is going to be coming soon in all these ageing, slow, useless old PCs replaced by high performance Athlon XP driven PCs.

Absolutely. Thanks very much, Jerry.

Our next question will come from Tim Mahon with Credit Suisse First Boston. Please go ahead with your question.

Yeah. Thanks. Jerry, can you tell us how much mobile units grew sequentially and how your relationship with

is on the integrated solution for mobile?

I think our mobile unit has got to more than 10% for the first time, which means there is something over 800,000 units, and some

gave me an indication that I am correct. They just pay up the 800,000 units and that

for the first time, I believe, were more than 10% of our total shipments.

has a very interesting product getting that out in time to match up with some of the design cycles has been a bit of a disappointment and I'm hopeful that we'll be able to recover some of the design opportunities that have been lost because

was a little late with their chipset.

if I may. Trying to go back to your Flash commentary, I think you said that you had 64 and 128 Megabit in backlog and that was going to help you drive your Flash revenue up sequentially. I am just curious. You said you were just sampling that. I would assume those products from MirrorBit.?

The MirrorBit product is coming in the second half, but we have non-MirrorBit

those densities. It did not become much more competitive in cost but, you know, we like to make money. You know, Memory has been a moneymaker for AMD almost always. It has been

product lines, you can surmise that with the precipitous

Memory revenues and the pricing pressures, Memory is currently not contributing to profitability. We expect that it will soon change that again.

I mean, if you were to take a step back and take a look at your end markets for Memory today, given that you have lost a little bit of market share probably in some of the cell phone guys. Can you give us a rough idea what that breakout would be?

Well. I don't have that at my fingertips. I think, our biggest issue was that we had some 40% of our revenues last year that came from the communications infrastructure business and that has dropped off precipitously. In supporting that market, we lost market share a big time to Intel and to cell phone space.

Right. And then just a last one. You keep mentioning Intel as your only real competitor in Flash but it would be apparent if you look at the data that

microelectronics has really gained a lot of share against you guys. How come you have discounted those guys as a competitor?

They don't have a follow on to compete with

Flash or Strata Flash's

MirrorBit. Without that they're last year's news.

. Thanks Jerry.

Our next question will come from Mark Edelstone of Morgan Stanley. Please go ahead with your question.

Yes. Thank you very much. This is Gary

for Mark. Good performance on the gross margins, gentlemen. Just wondering with Flash Memory doing better but microprocessor units seemingly down in the second quarter, what do you expect gross margins to be like in the second quarter, and for that matter, do you have an expectation for 2002 as an average?

I guess what I'd like to say there is since my Chief Financial Officer rightly so refuses the comment on gross margins going forward. Since, unlike our principle competitor, we don't enjoy pricing power in a monopoly position. Our gross profit margin is driven by top line. So, the gross margin will be dramatically different if we hit our 900

and it would be if we only achieve 820.

Do you have an expectation if you were to hit the mid point of your guidance, where the margin could be?

I'm sorry. If we were do what?

The mid point of your guidance range?

No. We're not going to comment on gross margins.

Okay.

The second question, you mentioned that shipments into distribution probably were up in the quarter?

A: No. I did not. You misheard me or you misunderstood me. I said I believed that since the total market shipment is the total shipments between the two major players and assuming some small contribution from

and/or whatever is left from

.

Right.

Forty one million units of processor shipments, that number was larger than what I believe the PC sales were. Some inventory was accumulated in the channel, not necessarily in distribution, it could be at the OEMs, could be at Best Buy, it could be anywhere. As a result, that inventory replenishment, I think added to the unit opportunities for both Intel and AMD in the first quarter and we don't see that situation in the current quarter. So, we believe that units will be tracking actual

.

Great.

You are

comfortable with the level of channel inventory?

Oh yeah. Very. No problem.

Okay.

It is not going to be a super charger for shipments. It is all about June. If business looks good by the end of this quarter, we expect to see, you know,

in demand in June.

Okay. Thank you very much, Jerry.

Our next question will come from Jonathan Joseph with Salomon Smith Barney. Please go ahead with your question.

With regard to your factor than expected acceleration of the

130 nanometer process sector, what kind of cost savings do you expect to realize on the Athlon dye?

At the

, Jonathan, the 130 nanometers kind of doubles roughly the number of dye available in a particular wafer and although, there is that added cost from the complexity of the technology, the doubling of the potential candidate is way more and significantly all right in the

additional cost in the technology.

So, we expect a significant improvement there. Also, the

what we call it time to money, the speed

, which we can achieve, yields in the technology change is pretty important in getting to cost down. We already are far exceeding at the current early level. The number

about the way through them, we have in the previous technology. So, we are very confident that it will continue.

With regard to the conversion of Fab 25 to Flash, will you be producing Duron in Dresden or will Duron

must go on at

?

Our plans are not to manufacture Duron in Dresden or

. Our plans are to be totally an Athlon based product line by the end of the year and at this point of time, that is all, I really want to say.

Yeah. I just like to add to that. As you know, the only real difference between Duron and Athlon is the size of the

cash. From the dye size and

point of view, with our 130 nanometer technology and going forward, we have the lead that the cost advantage that we get from going to copper among other things will enable us to produce Athlons competitively with the old aluminum Duron production such that we can offer the customer, higher performance at competitive cost and if anything, improve our margins.

The customer success is our success, so we are going to give a copper

full size L2 cash product at prices, which you'd be expecting to pay for a Duron product.

Great. Thanks.

In terms of demand in microprocessor, is the majority of the demand more at the 2000 or 2100 or less in that area?

I said the high end. I didn't say the very high end. Frankly, the very high end is quite low volume. The number of products that would be 2.2 GHz of Intel or model 2100 and 2200 from AMD, those are not the big volume. The big volumes are couple of speed grade down from that. Two speed grades down were the volume is. One or two speed grade down from the highest volume is where the volume is and that is where we saw the volume. And then the next coupled of speed grades, the so called the third position on the performance.. Mainstream sector it is called. And the upper two performance level and the value state.. we are just

much demand.

are looking for the best processor at the lowest possible cost for the low-end machine, or as I say that sweet spot which tends to be skewed higher.

Was your sweet spot for your whole portfolio products in the 1600-1700 range or little higher than that?

Little higher than that more in the 1800 plus range and moving up from there this quarter.

Our next question will come from

from JP Morgan.

Good afternoon gentleman. On the timing of the transition to 0.13-micron processor for Flash volume ramp up, could you comment on that?

We intent to begin limited sampling of the products at the end of this year of a 130-nanometer Flash. We'll have production next year.

Great! And then could you remind us of model number of the introductory product and volume though we manufactured .13 micron Athlon product?

Right now 2100+ is the product that we offer. We expect to ship 2200 model this quarter.

Also in the earlier statement, you mentioned that this year you will see some benefit from the

.13 micron process, could you elaborate on that a little bit more?

We expect to qualify our .13-micron Athlon product at the

sometime in the second half of this year and begin productions ramp. I expect that any production shipment at the

will probably occur in the early part of next year.

In other words you are anticipating that Fab 30 will not be able to provide enough capacity for the Athlon production.

No. I expect that our product demand in the market success with Hammer is going to put us in a position for Fab 30 will have dedicate significant resources to Hammer and we'll be able then to use

for a large number of our

.

Our next question will come from Thomas Thornhill from UBS Warburg.

On process for units, in the first quarter you said about 41 million in total for the market and some inventory build. What do you think PC unit consumption was so that the balance would be what went into inventory? And secondly, your comment on you think your PC units down 5 to 10 in Q2. What would be the distribution do you think across server notebook desktop? Do you expect them all to be down or desktop down, notebook up, something along those lines?

I think server will be up because I think it's a small market nearly 1-2 million units a quarter. So that is relatively small market. I think that there's a good opportunity for that to grow. The desktop definitely is not going to be up. I see that as a potential weak spot. Mobile should be good and workstations are

. So, I don't know that that's even material. Relative to the first part of your question,

would be something like 10 percent less than the units shipped, in the 36-37 million units range. I do not have better numbers than that. Those numbers should be out soon. As far as PC sales I just don't have a better numbers than that.

Thank you. Second question. Could you clarify some of the comments that were picked up in the press regarding Microsoft and Hammer and your testimony there as to Microsoft's support for Hammer? When do you think you will get a public comment on that since there has been one?

The press is trying to somehow tie my testimony as asking Bill for a favor. The realities are I was happy to testify for Bill. The timing was awful because as you probably know we have got a shareholders meeting next week, we had an earnings announcement this week. I was on my way Japan when he asked me. So, the favor wasn't testifying because I'm great believer that the windows platform has had a

benefit on the industry. I think that a common platform that companies build products in support of

win a good deal. I was very happy to do that. We've been working with Microsoft on our

extensions for years. We're not prepared to say when they'll make a public announcement. We're highly confident that they'll support us. We have given Microsoft working samples in platforms running code, and they're very pleased with the performance. So, Microsoft will have to make the determination of when they want to do whatever they want to do. But I am highly confident that they will be supportive of our processor.

Thanks for clarifying that.

Our next question will come from Larry

of Cantor

.

Thanks. Two questions. Can you give us some idea what the pricing on the mobile has been? Do you anticipate being able to get up to 2.2 GHz type products in the 2H of the year for Athlon?

The answer to the second question is absolutely. We have some significant headroom in Athlon. We're making modifications, as we speak as part of engineering, as part of evolution, not a question in my mind that we'll get to those levels of speed. More importantly we'll able to do things with the architecture to get the model numbers, which are more important than the gigahertz rating. Intel for the first time is now putting an A on the end of some of their gigahertz numbers. It implies that this one has a got a bigger cash than the other one's got and so it has got higher performance,

model numbers.

So, lets stop talking about Gigahertz. It is interesting, but it's

. We have countless application performance. We believe that we're going to be able to continue to increase the architectural performance of our device as we increase the clock speed. We expect Intel to do the same thing. It's a two-man race and we're going to be to and fro in going forward. The bottom line is customers like the improvement in performance and the ability to choose among two very good suppliers.

Regarding the pricing, we don't want to give you that kind of granularity. We think it's irrelevant. We have counted our total fixed cost which is the most important thing. The variable costs aren't much different whether they are mobile or desktop.

Thank you.

Our next question will come from Dan Scovel of Needham & Company.

Jerry earlier mentioned 80 percent of revenues from Athlon. Were you taking about processor revenues?

Yes. Obviously 80 percent of our revenues didn't come from microprocessors. 80 percent of our processor revenues nearly came from Athlon.

Obviously you don't want to answer the question about breakeven level or gross margin. So let met go for the ....

We said it is in the 900 million range and it is going down.

In the past you said that you guys expected to breakeven a better in second quarter. Are you still holding with those earlier comments?

I rather have you take the statement I just made. If we make the $900m of the range, we make money. If we make the low end of the range we won't make money.

I will reflect. Thank you.

Our next question will come from Ben

of Deutsche Banc.

Just want to clarify something first. Did I hear you at the beginning say that Flash units were up 12% QonQ and revenues were down 18% QonQ?

Yes. You got it right.

Do you think that just one quarter later prices can suddenly stabilized sequentially?

We have a richer mix of these multi chip packages and the higher density products, and prices have stopped going down.

Okay great! Second question I had, you either weren't sure or didn't want to communicate 2Q Flash expectations, which is easier for you to forecast in 2Q, is it units or ASPs?

We didn't say that. We just haven't put parameters on substantially. Obviously in order to get revenues up, you have to have an assumption on ASPs and an assumption on units. You also need an assumption on density. So, this is kind of a mix. You put it all together are we're content to merely say that revenues are going to be up substantially.

Great! Second question it is probably a bit difficult. Would you be prepared to make any comment on the rate of reduction in your breakeven level you can achieve through this year?

I don't think we should. I think that we had indicated at one point we were going to say we at 100 million a year by our cost reductions. That would've taken our breakeven point down by 25 million, when it was something over 900 million, and that'll get you at least in the right neighborhood. Yes, I like you had one other thought.

We're also spending record levels of R&D. We've reduced our breakeven level through cost reduction and cost control. We're spending on record levels. We're focused on a balance making sure that we're building for the future while trying to maintain breakeven levels at levels where we can show our profits and hopefully that will be soon.

Our next question will come from David Wu form

Once you start launching the 'Hammer' family, how would you position the two product line, the Hammer with the 64-bit extension and presumably a new architecture that will run 32-bit much better than the Athlon? How would you

the product transition beginning 4Q and then into first half of the next year?

You are going to really appreciate this. That is what is the magic about Hammer. Hammer is a unique concept and with a single buyoff, we can position it so that you can boot a 64-bit operating system or a 32-bit operating system. So, every Hammer product that we ship is 64-bit capable. But you can boot it, have a buyoff and just run

on Windows XP or you can boot at 64-bit operating system and a 64-bit capability. So, we're going to position the Sledgehammer against Itanium and high-end Xeons and we are going to position Clawhammer as our P4 killer.

Where would be the Athlon fit once the Clawhammer is shipped?

Well, it is all about process performance. Obviously, the Athlon will continue to be a viable product as long as it has performance, which is competitive at the price point. So, I would expect that Athlon will be contributing revenues through all of next year and beyond, particularly with the mobile opportunities.

Our next question will come from Brian

from Bear Stearns.

Just wanted to know I could get some additional color on your processor revenue mix, maybe by customer type OEM vs. distributor?

I don't think so. I think that you should know though that what you say by distributor, we do not record as revenues shipments to a distributor until they resell the product. So, whether the customer buys the product directly from us, or he buys it though a distributor, until it's actually sold to a user, we don't record that as a sale.

I think a better question would be how much of your business is in the US, and how much is in Europe and Asia. I think we're doing extremely well in North America, Europe and Japan. As I said earlier, we're not doing nearly as well against our competitor in China, Latin America, and India. In the first quarter, I think that is evident because 65 percent of our sales were International, which was down from 73 percent in the prior quarter. So, I think it's a good description where our business comes from.

Great! Thank you.

Our next question will be a follow-up question from Joseph Osha from Merrill Lynch.

Looking at the mobile market, do you think that there might be a potential for some market share gain there, that might protect

or cushion what's going on comparatively in the desktop?

In the short term, we have a performance advantage over Intel as you probably know. As a result of that we're going to see great things. I expect to see improvement. I expect to see progress. But unfortunately, the design cycles for mobile are longer than they are for consumer desktop. We announced today this 130 nanometer availability of outline XP for Mobil

. Some of the smaller guys can turn around faster. So, that's helping us. It'll take a little longer for the big guys to go with us. So, I would expect that, it would be more at the end of the year kind of benefit rather than anything in near term.

The 10 percent of your total mix that you shipped this quarter are obviously just 0.18 down. Am I right?

Not all because we made our initial shipments in the 130. They were in the tens of thousands and not in hundred of thousands.

Our final question will come from Doug Lee with Banc of America Securities.

Could you give us more color on the unit mix between Duron and Athlon in the quarter, you mentioned that Durons were a little more than last quarter and Athlons were a little less, but as far

absolute numbers was a sort of 50-50, 4m a piece?

No, it is more in the 4.5-4.7 million on the Athlon, and the balance was Duron.

Historically, the second quarter is very back-end loaded. Can you give some flavor based on the history, how back-end loaded is the second quarter? How much of process

do occur in the month of June on a traditional basis?

Most of them. More than 50%.

We see no further questions at this time. Please continue the presentation or any closing remarks.

Thank you everybody for participating in our call. I will remind everybody next is our shareholder meeting that will also be broadcast. Thank you for participating.

This is the 119th quarterly report that I had participated in as CEO of AMD. It has been a pleasure. That is 30 years as a public company and 119 quarterly reports. Next time, I won't be the host but I will be listening.

That concludes the conference call for today. We thank you for your participation.