美光科技 (MU) 2006 Q1 法說會逐字稿

Good afternoon.

My name is Jason, and I will be your conference facilitator today.

At this time, I would like to welcome everyone to the Micron Technology conference call.

All lines have been placed on mute to prevent any background noise.

After the speakers' remarks, there will be a question-and-answer period. (OPERATOR INSTRUCTIONS).

It is now my pleasure to turn the floor over to your host, Mr. Kipp Bedard.

Kipp, you may begin.

Thank you very much.

I would like to welcome everyone to Micron Technology's first-quarter fiscal year 2006 financial release conference call.

On the call today are Mr. Steve Appleton, Chairman, CEO and President;

Mr. Bill Stover, Vice President-Finance and Chief Financial Officer; and Mr. Mike Sadler, Vice President of Worldwide Sales.

This conference call, including audio and slides, are available on Micron's home page on the Internet at micron.com.

If you have not had an opportunity to review the first-quarter 2006 financial press release, it is available, again, on our website at micron.com.

Our call will be approximately 60 minutes in length.

There will be a taped audio replay of this call available later this evening at 5:30 PM Mountain Standard Time.

You can dial into that by punching 973-341-3080, using a confirmation code of 6813208.

This replay will run through Wednesday, December 28, 2005, at 5:30 PM Mountain Standard Time.

A webcast replay will be available on the Company's website until December 21, 2006.

We encourage you to monitor our website at micron.com throughout the quarter for the most current information on the Company, including information on the various financial conferences that we will be attending.

During the course of this call, we may make projections or other forward-looking statements regarding future events or the future financial performance of the Company and the industry.

We wish to caution you that such statements are predictions and that the actual events or results may differ materially.

We refer you to the documents the Company files on a consolidated basis from time to time with the Securities and Exchange Commission, specifically the Company's most recent Form 10-K and Form 10-Q.

These documents contain and identify important factors that could cause the actual results for the Company on a consolidated basis to differ materially from those contained in our projections or forward-looking statements.

These certain factors can be found on the Company's website.

Although we believe that the expectations reflected in the forward-looking statements are reasonable, we cannot guarantee future results, levels of activity, performance or achievements.

We are under no duty to update any of the forward-looking statements after the date of the presentation to conform these statements to actual results.

With that, I would like to turn the call over to Mr. Bill Stover.

Thanks, Kipp.

For our first quarter, net sales totaled $1.36 billion, and the Company recorded net income of 63 million or $0.09 per diluted share.

The diluted earnings per share calculation is based on 707 million shares.

That's inclusive of the 54 million shares underlying our convertible debt.

It should be evident from this level of net income that our diversification efforts are progressing meaningfully.

These financials do not include any results of IM Flash Technologies, our joint venture with Intel, which will begin operations in January.

Let me take a few moments to talk about how IM Flash Technologies will impact our financials.

First, our joint analysis with Intel confirmed that the venture will appropriately be consolidated in Micron's financials.

Beginning in a few weeks, our results will reflect consolidation of the joint venture's manufacture of NAND Flash memory.

Remembering that both Micron and Intel will take approximately half of the output, the net sales line will pick up the NAND production sold to Intel at manufacturing cost, and the net sales line will pick up the other half of the production as it is sold by Micron at market prices.

Cost of goods sold will reflect the cost of manufacturing NAND devices.

The partnering with Intel recognizes the significant value in the investments Micron has been making in research and development.

Fundamental to the partnership is the sale of process technology designs and know-how for in excess of $200 million.

The technology sale is expected to be recognized as a gain reflected in other operating income.

On an ongoing basis, our consolidated research and development expenditures will be reduced by approximately $25 million per quarter as a result of the cost-sharing with Intel.

The sale of technology to Intel and the anticipated $250 million prepaid from Apple will bolster our consolidated cash and investment position by approximately $500 million in January.

With formation of the venture, Micron's consolidated capital expenditures will rise to approximately 2 billion, approximately a $500 million increase for fiscal year '06 over the number which we previously communicated.

Overall net sales were up 8% over the immediately preceding quarter.

Gross margin for the quarter came in at 23%.

The slight increase in gross margin percentage is quite impressive, actually, given the price pressure in commodity PC DRAMs and is a testament to the benefits being derived from our diversification efforts.

Gross margin on tech semiconductor products was slightly higher than the overall reported gross margin for the first quarter.

Research and development expense for the first quarter was 166 million, slightly higher than the fiscal 2005 run rate.

A benefit of our partnering in NAND Flash will be the sharing of certain R&D costs.

We expect the run rate for consolidated R&D expense to reduce by approximately $100 million per year through our partnering arrangement.

As always, future R&D expenses will vary significantly with the number of wafers dedicated to new device development and qualification.

Selling, general and administrative expenses stayed in the 85 to $90 million band throughout fiscal 2005.

We expect the quarterly run rate in 2006 with the consolidation of IM Flash Technologies to be between 95 and 100 million.

For the last three years, we have had continued improvement in cash flow provided by operations.

The overall improvement reflects our successful diversification into specialty DRAM and CMOS image sensors.

The first quarter's operating cash flow reached in excess of 425 million, in part benefiting from successful reductions in inventory levels.

Our finished goods inventories decreased by 22% quarter over quarter.

As of the first quarter end, Micron had cash and investment balances approximating $1.4 billion.

Total debt declined to 1.1 billion, and call protection on our $632 million convertible notes expires in February 2006.

The notes due in 2010 are convertible into common stock at a conversion price of $11.79 per share.

One other item -- this quarter is the first period reflecting stock option expense under Statement of Financial Accounting Standards 123R.

The total period expense was $3.8 million.

With that, I'll turn the commentary over to Mike.

Thanks, Bill.

We continue to be pleased with the robust demand environment in the mobile communications, consumer electronics and computing markets.

These markets account for the vast majority of demand for Micron's semiconductor memory and imaging products.

As evidenced by the quarter-over-quarter 11% Mb sales increase and corresponding finished goods inventory reduction, as referenced by Bill, the markets are healthy and easily absorbing the output from our production facilities.

Micron remains focused on achieving an optimal balance of business in the key end markets.

We have had another quarter of solid execution towards achievement of that objective.

While we have significant revenue exposure in the world's largest market for semiconductors, computers, we believe that more rapid growth in communications, consumer and industrial markets will drive more value.

With this in mind, we continue to add to a product portfolio that strengthens Micron's value proposition to customers in all of these market segments.

While the computer market demand is solid from both unit growth and memory content standpoints, the industry has grown DRAM supply such that price pressure on commodity chips used by notebook and desktop PCs has intensified during the quarter.

Micron is not immune to this price pressure.

We believe, however, that supply and demand will ultimately find a balance.

We expect that our ability to reduce costs through yield improvement and technology advancement will track market price declines.

While we are not expanding share in the notebook/desktop area, these segments are and will continue to be significant demand drivers both for the industry and for Micron.

In the computing arena, we are focused on being early to market with high-density chips and modules.

Good execution is driving more segment share in workstation and server applications.

Our 1 Gb and 2 Gb DDR and DDR 2 components and the 2, 4 and 8 Gb modules that they enable are current examples of solutions that create value for Micron and our customers.

The competitive playing field in the high-reliability, high-density memory area is not solely dependent on selling price per bit.

With a focus on time to market and technical collaboration, we are able to create more high-end computing business for Micron DRAMs.

This shields us somewhat from the severe price competition that results from an oversupplied commodity DRAM market.

Mobile phone terminals are among the most exciting applications for Micron, and we now have four product categories targeting this space, including image sensors, cellular RAM, mobile DRAM and both embedded and external NAND Flash.

To put it simply, the market here is growing, we are growing share with existing products and we're introducing new products that are capturing new design wins for us.

Memory content in mobile phones is expanding more rapidly than in any other high-volume electronic product.

NAND Flash, cellular RAM and low-power DRAM products play a critical role in the memory subsystem.

A sweet spot memory solution in a high-end phone today is a multichip packaged product built with a 256 Mb NOR chip plus a 64 Mb cellular RAM.

This is slowly transitioning to 512 Mb NOR plus 128 cellular RAM or low-power DRAM.

As we move through 2006, we believe the market will begin to shift to MCPs built with 1 Gb NAND chips and 256 Mb or 512 Mb low-power DRAMs.

We currently play a key role in the mobile phone memory market with our family of cellular RAM and low-power DRAM products.

Our hope is that this becomes a commanding role with both the NAND and DRAM pieces in play.

Our recently announced NAND Flash manufacturing venture with Intel will help us achieve scale and improved time to market with NAND devices for the cellphone market.

With our product lineup, technology and infrastructure, we couldn't be more pleased with our position in the cellphone memory ecosystem.

I think by now, you are aware of Micron's position atop the CMOS image sensor market.

While camera penetration rates are approaching 60% worldwide, we're continuing to outgrow the market and have now even captured multiple design wins in the Japanese market, the last frontier for CCD sensors.

We're introducing cost-reduced versions of VGA and 1 megapixel sensors, as well as new 2 and 3 megapixel sensors for high-resolution mobile phone cameras.

Our current mix of revenues in the camera phone space is approximately one-half VGA and one-half 1 megapixel and above.

The VGA sensors are having a longer-than-anticipated live as camera penetration reached the low end and many phones now even feature dual cameras.

We were expecting that market demand for our sensors might temper somewhat seasonally, but this is proving not to be the case.

We are moving some memory production within the micron fab network to create even more near-term imaging capacity to seize the market opportunity.

Camera-enabled mobile phones have been the primary driver of revenues for our CMOS imagers, and this will continue to be the case as we move forward.

While not the most significant revenue generators today, we are devoting a number of product development and field resources to other promising markets in the imaging area.

Some examples would be consumer items like digital still cameras, PC cameras, and medical and automotive applications.

Each of these product areas is of interest not only from the imaging perspective, as we also have the capability of providing application-specific memory devices into these markets for our common customer and application base.

Our strategy of developing a diverse set of semiconductor memory and imaging products is clearly paying financial dividends.

From our perspective, the real value is coming from strengthening and broadening customer relationships.

These relationships have created a solid foundation from which to build upon.

Thanks for your continued interest and support in the Company, and I'll turn it back to Kipp.

Thank you, Mike.

Right now, we'd like to go ahead and take questions from callers.

Just a reminder -- if you are using a speakerphone, please pick up the handset when asking a question so we can hear you clearly.

With that, we'd like to open up the lines to questions.

(OPERATOR INSTRUCTIONS).

Randy Abrams, CSFB.

I first had a couple questions on the CapEx guidance.

Could you talk about the 1.5 billion Micron spending and 500 million for IM Flash in terms of capacity additions on 300 mm versus back end and shrinks?

Well, just initial clarification is we had been guiding fiscal year 2006 at 1.5 billion.

And with the consolidation of IM Flash, there's approximately $500 million of joint venture spending which will be consolidated within the fiscal year 2006.

The 1 to 5 billion reference for Micron is still representative of our cash flow as the additional spending at the venture is funded by the initial capital coming from Intel and partners.

And then maybe the mix between 300 mm and shrinks and back end, how that splits out?

Well, almost the vast majority of new fab spending is 300 mm.

And then, of course, we don't really break out between what goes to shrink versus just new capacity in this environment because remember that we, even at this point in time, we have been finishing out the RAM and MTV in a quarter of this, and of course that will start increasing as we move to the rest of this capacity through the year.

So I think it's fair to say that the vast majority of it is on the front end.

And in our particular case, I don't have the detail and breakout between new capacity and what is advancing the process, because the joint venture obviously impacts that, and I just don't have that detail at the tip of my fingers.

And maybe looking at the near term just on bit production growth, could you talk about what you are targeting and maybe how that splits out between DRAM and non-DRAM?

We're looking at mid- to high-single-digit production bit growth.

And of course, that's always exclusive of the image sensor ramps.

And then your finished goods inventory looks like it came down in the quarter.

I guess, into spending (ph) or looking at your growth, do you expect an inventory build when you look at the next few months?

If we were going to anticipate an inventory build, this is the quarter that we would do it because of the seasonality as well as the various holidays that we have with, of course, our holidays here in the States as well as Chinese New Year.

In our internal projections, we have built in a slight inventory build.

Based on the strength of the market today, it's plenty strong to be absorbing everything that we're producing.

Tim Luke, Lehman Brothers.

I was wondering if you guys could clarify what you had said with respect to seasonality and how it may have differed in terms of your expectations.

And I was also just wondering if you could clarify your comment with respect to R&D, which has been running at the 166 level.

And it sounds like you are anticipating that that would now move lower with the collaboration with Intel.

This is Mike.

I'll speak about the seasonality.

First of all, on the quarter we just completed, from a demand standpoint it really played out just about like what we expected.

All the end products were quite strong with respect to demand.

We did, however, in the commodity DRAM area, see pretty strong supply growth, which put a lot of price pressure on the commodity DRAM products that are being used in both the notebook and desktop PCs.

But demand tracked about like we expected.

Now, typically, in a calendar Q1 type of scenario, which roughly corresponds with our fiscal Q2, we would expect DRAM demand from PCs to be flat to, say, down 5%.

And on a recent polling of our customers for this time period for the coming quarter, if we add everything up, it's about flat.

So flat quarter over quarter from a seasonality perspective into fiscal Q2 for us.

Does that address your question on the seasonality piece?

That's very helpful, yes.

On the R&D front --

(multiple speakers)

I'm sorry; go ahead.

I think maybe there was a second part of that you were about to say -- the seasonality in the image sensor business, for example.

Yes, thanks for reminding me.

We would have expected -- again, the primary driver for our image sensors today is mobile phone cameras.

And we would certainly have expected some seasonality there as well.

And actually, we would not have been surprised had demand kind of slacked off a little bit once we got past the Christmas selling season.

But that was not the case at all.

Our demand for our image sensors is absolutely growing into the current quarter.

We've got customers coming in here just about every day to pound us for more supply.

So where would have expected maybe a slight breather from a seasonality standpoint on the image sensor side, things are zooming from a demand standpoint.

Really, there appears to be no end in sight in the near-to-intermediate term for growth there.

And on the Flash side as well, again, this is a consumer item; the NAND Flash is a consumer item.

We would have expected -- there may even be some seasonal slowdown in NAND Flash demand.

But my view of things is that demand today so greatly exceeds the industry's capability to supply that even a seasonal slowdown is not going to have a material impact on the current imbalance of supply and demand.

On the R&D front, with the question about what we expect to happen there, clearly we're going to get some efficiencies from cost-sharing.

But I also think that we need to separate in terms of short term and long term.

Long term, we think the R&D will be much more efficient with us jointly sharing the cost of that.

But also keep in mind that in the shorter term, we have a much broader portfolio that we're going to be pursuing, and we're going to have some activity to try to bring the innovation from both companies to the forefront.

So when you think about it, we've got new fabs and we've got an expanded product portfolio now that we have this significant capacity that we'll be taking advantage of.

In that scenario, at least in the short term, we really don't expect it's going to go down.

We're going to get some benefits from the cost-sharing, but -- anyhow, Bill, do you want to add a comment?

Steve's comments lead us right to a range of about 150 to 170 million, where in the near quarter, we're going to be toward the upper end of that range, and a couple quarters out we would expect to be towards the bottom of that range.

That makes sense.

Any color just with respect to pricing and how that may play out in terms of gross margin outlook?

Very difficult to project what's going to happen with respect to pricing.

If you paid attention to some of the news that's been out in the public on spot market pricing in the DRAM area just in the past week or so, it appears to have stabilized at a level that's much lower than we would have hoped for.

But it appears to have stabilized.

Our contract renegotiations that occurred midmonth with our big OEMs resulted in flat pricing.

So it appears that we're through the storm, anyway, on the strong price reductions that we have seen in the DRAM area.

And on the NAND Flash area, there's really not much price pressure at all.

Prices are relatively stable.

In the CMOS image sensor area, we are kind of in a sole-source situation with virtually all of our customers.

So there's not a lot of commodity-type price pressure there, either.

Adam Parker, Sanford Bernstein.

What percent of your revenue within that 45% you reported -- can you break that down into CMOS, NAND and specialty DRAM for us?

Well, I can give you -- we prefer to stay away from the revenue piece of that.

But again, from a wafer allocation standpoint, we are running right around 50% of our product in kind of the more PC-oriented DRAM capacity, 30% in the specialty range, 25 to 30% in specialty.

And then that leaves you 15 to low 20% in everything else.

So a combination of Flash in general and image sensors in other.

So last quarter, I think you guys said that NAND was kind of in that 6 to 8 range, and maybe CMOS rounded up to 10.

Did the CMOS outgrow the NAND sequentially here in the November quarter?

Yes.

Materially, I assume?

Yes.

Okay.

And then what percentage of your DRAM revenue came from tech during the quarter?

Or again -- or do you not -- want to avoid revenue commentary?

Well, like we have stated in previous public disclosures, it runs about half of our DRAM, commodity DRAM output.

And do you think there's any reason that would change, Kipp, in the coming few quarters?

No, I don't see any change at all until, as Steve has mentioned on prior calls, tech begins to go through some type of a 300 mm transition, which we've spoke to publicly being towards the end of '06 or '07.

Last question is on the options accounting.

It seems -- two things.

One, can you allocate that across R&D, SG&A and COGS for us?

And then also, it seems like it came in a little bit below what you had previously expected.

Can you explain that modest reduction?

Yes.

That $3.8 million, probably half of it in the cost of goods sold arena; another 30%-ish in R&D and 20% in SG&A is a rough breakout for you.

Will you give exact breakouts in your Q, or--?

Hadn't gotten to that analysis; did not expect to.

As you appreciate, $3.8 million is not significant to the overall results of operations.

The estimate for the year is still running in the $25 million range, and the fact that the first quarter was somewhat lower is just a timing of grants issue.

So, still -- so we should assume 25 million at what price in your stock, sorry?

It really has, right now, Adam, has more to do with the timing of the grants, more so than a point-in-time stock price.

So it still going to be around 25 million?

Correct.

Doug Freedman, American Technology Research.

Could you guys rank the products by gross margin?

You have offered that in the past.

Sure, we could do that for you, Doug.

Like prior quarters, image sensors continues to lead the parade.

Interestingly enough, specialty DRAM and NAND was pretty close in margins this particular quarter, and then, of course, followed up by DRAM gross margin.

Just looking at my numbers here, did specialty DRAM grow or did it contract in the quarter?

It did grow.

It did grow single digits.

It grew single digits.

Okay.

And then what should we look at as far as the NAND ramp?

I was a little surprised; it appears that NAND did not grow nearly as fast as I was thinking it was going to.

And any sort of color you can offer on sort of the aggressiveness at which we can expect NAND output to start ramping?

You bet.

As we've previously stated, we will be fairly flat in wafer allocation for NAND, until we can start making some adjustments in Boise and we start seeing output from Virginia, which is later this year.

So expect the NAND Flash, again, ramp to be fairly flat for the next couple of quarters, and then we start to see some ramp as we get into the latter part of summer into the end of the year.

And then can you offer a bit shipment growth for the quarter, because it appears as though you worked down quite a bit of inventory?

Looking at my model, I'm coming up with the fact that you guys might have shipped as much as 27% bit growth.

Can you enlighten us as to sort of whether you guys feel like you took some share during this period, or was this really driven by the handset market consuming quite a bit more?

Let me start with the first part of your question.

Then I'll toss it to Mike, see if he wants to comment on some of the segments.

But in terms of guidance for production bit grow, we'll be mid- to high single digits this quarter.

And again, that's exclusive of image sensors.

And again, that's production.

And Mike, would you like to make any commentary on different segments?

Sure.

First of all, I'm not sure how you got your 27% calculation or whether it was back of the envelope or not.

But we did -- last quarter, we increased bid shipments by 11%, I think.

I think that's the correct figure.

Yes.

And that was against a production increase of 8%.

So we did draw down inventory somewhat in the quarter.

Where we are seeing strength or where from an application standpoint where we are seeing more dramatic growth of course in handsets, both from a memory content, just a sheer demand growth standpoint as well as our market share growth in there.

So handsets would probably be at the top of the charts with respect to growth.

Servers, also a significant growth area for us.

And we have given up some share, although I don't think it's significant.

But we have given up some share in the notebook and desktop space.

John Lau, Jefferies & Co.

Looking toward the future memory architecture in the cellphones, you had mentioned that you had four products that play into the cellphone area.

Can you tell us what do you think the architecture is going to be?

Is the NAND still going to be separate?

Is it going to be integrated into the cellphones, and how Micron would play in those different markets.

Thank you.

Yes, sure.

Both.

Today, our view is that NAND does not play a significant role in cellphone architecture.

Certainly, from an embedded standpoint, it does, from an external or a card standpoint.

Our plan is to participate on both sides of that.

One, our first MCP product with a NAND Flash device is going to be a 1 Gb, 1.8 volt NAND chip packaged with 512 Mb or a 256 Mb low-power DRAM product.

That is probably going to be a market reality in the second half of 2006.

So that's basically the embedded play for us.

And that's going to grow, of course, to 2 Gb NANDs plus 1 Gb DRAMs and so on and so forth.

And then the other side of that would be the embedded piece, and that would be through a variety of different card -- form factors.

And we supply chips to card manufacturers that get us in that way, and we are contemplating a rollout of other means of entering that piece of the market as well.

But suffice it to say that both the embedded and the internal pieces are of significant interest for us.

Thanks, Mike.

You mentioned an interesting comment that the NAND Flash in terms of your bit grow production is going to be kind of flattish for the next several quarters.

I was wondering if you can elaborate on that and if you are going to be participating or missing out on the market as you would try to adjust your capacity?

Well, we are -- as we have mentioned before, we are ramping NAND.

But it's predominately today in the Boise facility.

And keep in mind, in the Boise facility, we are also producing imagers, which are in strong demand.

We're producing the cellular RAM, which is in very strong demand.

In fact, really what I would think of it is kind of the first custom-designed, low-power DRAM device, is running in Boise for the cellular market as well, as an extension of the cellular RAM business.

So we have some challenges on balancing the capacity among some pretty strong products.

And as a result, the Boise will continue to ramp on NAND, but it's not as aggressive as you might expect it to be, given the strength in the NAND market.

But that's also because it's against other products that are doing quite well.

And so it's more of a calculated change in the balancing of the wafer resources as opposed to something dramatic.

That obviously will start to shift pretty heavily as Virginia comes online.

Really we are thinking in terms of Virginia, we'll start laying some wafers there in mid-'06, and then later '06 for Lehigh.

So both of those will be coming on, but really the next couple of quarters, as Kipp mentioned -- on a relative basis it's still growing for us, but on a relative basis compared to what will happen in second half of '06, it will appear to you to be somewhat flat.

Shawn Webster, JP Morgan.

I was wondering if you could provide a little bit more color around your pricing in the DRAM segment and I guess maybe in the sensors as well on an average basis, like what your DRAM pricing did quarter over quarter.

And then I have a follow-up, please.

Sure.

If we look at the commodity piece of the market, which, again, as Kipp mentioned, that's roughly half of our business, I think quarter over quarter prices were down around 15%-ish, I think, if I'm not mistaken, maybe slightly higher than that.

If you look at our overall DRAM portfolio, our price per bit was down about 5%.

So I think that speaks to the strength of the portfolio.

Essentially, there is heavy price pressure on the commodity area and pricing was kind of flattish for all the other DRAMs in the portfolio.

On image sensors, it's really a function of what we are doing with the product line as opposed to what the external market pressures are, because frankly speaking, for us, there are not a lot of external market pressures.

We're the big player in the market, and so in other words, we could take -- we did take average selling prices down quarter over quarter because we significantly grew our VGA sensor output in response to market demand.

And of course, VGA sensors sell at a lower price than 1 megapixel or 2 megapixel or 3 megapixel sensors.

So essentially, not a whole lot of price pressure there, just a function of what we are doing with the product portfolio.

And I guess two kind of unrelated questions.

The IM Flash joint venture -- is that signed, or do you expect it to be signed in January?

Yes, we expect to close it in early January.

And actually, that's all I have.

Thank you.

Jim Covello, Goldman Sachs.

A couple of quick questions.

Firstly, I just wanted to make sure I'm thinking about modeling the profitability of the NAND business properly.

The wafer output growth is going to be flattish for the next few quarters, and you're going to give up -- you are sharing half of the profitability with Intel.

So the NAND profits are going to go down for the next few quarters, until the Intel -- until the incremental capacity comes online, in which case you're obviously planning on growing it much faster than what you're going to share with Intel.

Am I thinking about that wrong, or do I have that right?

I think in general, that's right.

What we said was that for the next two quarters that the output would seem relatively flat, although it is still growing.

Really, it's the third and fourth quarter of '06 where I think we will start to see something of substance in terms of change.

And yes, we will be splitting the output.

And hence, we will be not getting whatever margin is on that half that runs through our partner, Intel.

But of course, the JV itself and its sale of the product to Intel will flow through Micron, and you've got to remember that the R&D benefit that we are receiving will be countering that.

So when you net that out, if you can help us from a modeling perspective, for the next couple of quarters, where there isn't any incremental output and you are sharing some of the profits, can you help us think about what kind of EPS impact that would be or something along those lines for modeling purposes?

No, not at this time, sorry.

Next question.

You talk about the DRAM industry or commodity DRAM finding a supply/demand balance -- I think that was in the opening -- or that it will find a supply/demand balance.

That was in the opening comments.

But hasn't the problem really been on the supply side from your competitors and not really on the demand side?

The demand for PC units has been great; it's been the supply issue.

What fixes it, if the Taiwanese DRAM companies haven't shown a willingness to scale back on their supply?

In terms of -- and particularly the historical impact, Jim, clearly, supply-side economics have had a greater impact than demand because the demand has been relatively good year after year in terms of its bit consumption, ranging from 40% or old days of 75-80%.

So your question about what will rationalize the DRAM supply going forward, we don't necessarily have a good answer because we don't have visibility to what some of our competitors are doing, specifically with respect to DRAM, into the PC space.

But of course, if we had the absolute answer to that, we would have been diversifying the portfolio.

Or maybe we would have been doing it slightly different.

But that's the reason we are trying to be exposed less to that product that goes right into the desktop space.

And I've heard both sides of the fence in terms of what the industry thinks is going to happen on supply.

Clearly, there's an allocation into NAND or, in our case, into NAND and DRAM -- or I'm sorry, NAND and imaging, and what I think of as specialty DRAM, because those markets are growing.

And others are trying to do -- some others, anyway, are trying to do similar things.

So that's what's hard for us to get a handle on is how much migrates, how much current capacity migrates out.

And then the new capacity you guys can go out and look at the models out there and what they are going to do with it.

But --

If I could drill down on that just for one second -- this will be my last one -- if I think about '05, we had great PC unit growth, we had a lot of capacity allocated over from DRAM to NAND and the image sensors, in your case.

And the industry pricing was still a lot worse than expected.

What kind of scenario do we need to see where we get better-than-expected pricing?

I'm not really sure that we have seen the result of a major effort on the part of those manufacturers that can shift capacity towards NAND.

There's not a lot that we can do with our capacity, but we are doing what we can in order to realize more NAND output sooner as opposed to later and sacrifice some DRAM.

There are two other big players in both the DRAM and the NAND area that claim they have the capability to shift capacity from DRAM to NAND.

And I think now is about the time when we should really start to see the results of that, if in fact that effort will bear fruit, because the NAND market is clearly undersupplied right now and the DRAM market is clearly oversupplied.

Those two guys would argue that they moved whole fabs worth of capacity over in '05, and they will do that again in '06, but no more.

Do you think differently about that?

I don't know -- no idea what their mindset may be.

If I were in their shoes and I had the ability to be as flexible as they've indicated that they do, I'd certainly be moving much more capacity towards NAND and away from DRAM in this kind of environment.

If you look at the margin differential, it's staggering.

David Wong, A.G. Edwards.

Just a clarification on the R&D.

All spending by the joint venture actually is consolidated on your income statement, am I correct?

So your guidance, when you said 170 million going down to 150 million, that's what shows on your income statement line?

Or it's more than that, because you were including payments by other third parties?

You've got it correct in that the consolidation of IM Flash technologies will include the R&D spending at the joint venture with our R&D spending.

And as a result of the sharing, it will be in the 150 to 170 range, even with the diversification and the additional product efforts that Steve outlined.

Glen Yeung, Citigroup.

I wanted to ask a little bit about the price reductions, cost reductions that you're seeing now.

I assume you're staying on your sort of typical path.

And looking at where spot prices are today, my second question is, is the difference between spot price and your cash costs at or better or worse than what it was at the beginning of last quarter?

We are not going to get into that kind of specifics.

If you'd like to try the question in a different manner, we might be able to help.

I don't know if I can, because I guess we all know what the answer to that -- kind of guesstimating (ph).

All right, well, let's maybe look somewhere else.

We are looking out at the image sensor market, and we're seeing a lot of strength, obviously, from handsets in general.

But we're starting to hear more and more talk about image sensor demand coming out of China, in particular.

And wanted to get a sense from you what you see from that market and what you think that does or doesn't do to the sustainability of image sensor growth -- strength.

Actually, we are not hearing much at all about silicon coming out of China for image sensors.

They are putting in capacity, more capacity to do things on the back end, if you will -- modules or CFA or something of that nature.

And maybe one or two of the foundries might try to do something there.

But we are not aware of, really, any silicon coming out of China right now.

I'm actually thinking of it the other way, in terms of demand starting to pick up more aggressively there.

On the demand side, actually, the camera phone market in China is -- the penetration rate is relatively low, and the end demand from the Chinese market is -- I would characterize it as insignificant today.

So that's a huge, huge potential growth area for us.

What demand growth we're seeing from China is strictly -- today, anyway, is strictly from an assembly standpoint.

So most of -- actually, if I'm not mistaken, virtually all of our end customers have, either directly themselves or through a subcontractor arrangement, a camera module assembly operation in China.

And that's why we're seeing demand coming out of China; it's primarily just that's where the assembly is taking place.

One other question, which is thinking about looking at the end of this fiscal year from now.

How do you think we can expect to see the rates of -- well, not the rates, but how can we think about the mix of your business -- and I guess on a wafer-out basis is a good way to think about this -- for NAND, CMOS, specialty DRAM and commodity DRAM?

We expect the commodity PC DRAM to continue to trend down.

We expect specialty DRAM to continue to moderate up.

And of course, NAND Flash will be beginning its pretty good ramp at that period of time, so it will see some pretty significant increases as well.

And image sensors continue to look very strong.

You still don't want to quantify that, though, I assume?

I'm sorry?

You can't quantify that in terms of what percentage that might make up at the end of the year versus now?

It's too far away.

We're going to continue to be opportunistic and move wafers around to where we see strategic benefit to do that.

And anything more than a cycle time gives us a pretty good opportunity to move wafers.

So it's a little hard for us to look out six, nine, 12 months and give you that ratio.

Kipp, is there a range or a level at which you just can't lower your DRAM, commodity DRAM, anymore because you've got certain responsibilities to OEMs, for example?

Is there a level you have to just keep producing at?

Well, I think that the responsibilities to the OEMs in the PC market -- it's obviously a slightly different responsibility to products where we are sole source.

So it's not quite that issue, so long as if we're changing our capacity around, as long as we give sufficient notice to the customer base, I don't think it's as big a deal on the PC side.

The challenge that -- well, one of the challenges that we have is we've already made big moves in capacity allocation.

Obviously, as Kipp mentioned, about half of it -- we are down from what used to be 100% a couple years ago or a few years ago down to about half of it now going into the desktop space, if you will, on the DRAM side.

And so we have capacity in place that's geared towards doing certain things.

We can migrate it over time, but when you start talking about those kinds of changes in a relatively short period of time, we run into some limitations that exist at least for maybe two or three quarters before we can substantially change that.

And an example would be we've put in significant capacity in 12 inch to really think of it (ph) as the high-volume DRAM products.

And if we're going to change that to the degree that we want to change that, it just takes us a little bit of time to do.

And so there's what you consider to be flexible capacity that you can change within the cycle time that Kipp spoke about, and then there are some others that take two or three quarters to try to change.

We still have some flexible capacity left, but we're going to run up against our ability to do something in one cycle time being limited.

And we have to make more wholesale changes on a facility basis.

Krishna Shankar, JMP Securities.

Of the 55% which is PC DRAM, can you give us some estimate of within that is the server workstation part of that joint portion, and can you talk about the profitability of that piece of the computer DRAM puzzle versus PC DRAM?

I'm not going to provide specifics on what portion of that would be the server piece, but it's less than half of that PC DRAM piece would be servers.

And for us, the server space is more profitable than the desktop and notebook space, primarily because it is a higher density, requires a higher reliability devices and modules and so forth.

So it's more profitable for us, and it is a growing piece for us as well.

And is there any consideration being given to just moving more PC DRAM, some of your older 8 inch PC DRAM fabs over into the NAND Flash joint venture?

Because the revenue per wafer is higher there, so why not even lower the exposure to PC DRAM even more?

Actually, we've been on that migration path.

That's what we're experiencing today.

We've continued to move PC DRAM into other products.

But the other thing to make note of is we don't have a lot of 8 inch capacity left running what you would think of as PC DRAM.

We have what we call a legacy DRAM portfolio.

We have obviously got the NAND we spoke about, and we have got imaging and some of the cellular-type stuff, which is consuming the vast majority of the 8 inch.

And the only exception to that, by the way, would be our partner in Singapore, TECH.

And they are on a migration path to 12 inch.

As Kipp mentioned, they are going to start doing that, actually some work on it here in the next quarter or two and then start that progression of change.

So I think what you have left after that is we'll have Virginia running 12 inch.

So there's just not a lot of 8 inch running what we’re characterizing as PC DRAM.

And then turning to NAND Flash, what steps are you taking to really start to expedite the ramp both in Virginia and Lehigh, just given the tremendous profit potential there?

How can you expedite the ramp of the NAND Flash joint venture manufacturing?

Well, we are aggressively as moving as we can.

It's kind of like having a child; there is a certain period of time that you've got to gestate, and it's just going to take that long.

So, to the extent that we can accelerate it, we will.

But there's only so many things you can do.

You've got equivalent supply times, etc., etc.

And can you talk about your product roadmap in NAND Flash?

I guess, right now, everything is on 8 inch 90 nanometer.

Can you talk about both the product roadmap in terms of density and line width and modular cell flash (ph) technology?

Sure.

We have been producing and shipping, obviously, since day one last December, so almost a year now anniversary, 2 Gb.

We have been sampling now a 2 and 4 Gb.

And, by the way, that was all at 90 nanometer.

We are now sampling 2 and 4 Gb at 72 nanometer as well.

And of course, we now have a product roadmap beyond that we have not shared publicly yet.

And how do you see pricing trends between the 2, 4 and 8 Gb NAND Flash out there, even though I realize you don't participate in 8 Gb?

What do you see in terms of pricing?

Actually, we do participate in 8 Gb by stacking four of our 2 Gb chips.

And from a market pricing standpoint, basically the market is at roughly parity on a per-bit basis for either a 2, 4 or an 8 Gb solution.

Paul Leming, Soleil Securities.

What did wafer outs average for the quarter, and could you give us a guesstimate as to what they will average for the current quarter?

Mid-60,000 wafer outs a week, and they will just be up very slightly in fiscal Q2.

Nimal Vallipuram, Benchmark Company (sic).

I have two questions.

The first one is that, Bill, you indicated that you would give some color on how the accounting is going to work with the new venture with Intel.

You said that half of it will go to Intel at a certain price, and the other half we will be able to sell it in the market price.

But the total cost of goods sold will go to the cost of goods sold -- all 100% cost of goods sold will go to your cost of goods sold.

Now, can you add any more color as to whether the Intel pricing is always going to be approximately below the market pricing at a certain discount?

Or, if the market price goes down significantly, would you have some sort of upside on Intel pricing?

This is Steve.

The basic way the JV works is that we are buying -- both of us are buying, equally, the product from the joint venture cost.

So there's nothing going on at the JV level besides that.

And whoever gets consolidated will just incorporate that concept.

I understand.

I guess my second question -- this has been asked by a number of people, so please don't -- I just want to get some idea.

When you exit in calendar 2006, can you give us any idea on a basis -- I think probably a bit basis would be the best way to look at it -- at least approximately what percentage of your bit output would you like or would like to be in NAND Flash as opposed to everything else?

Well, you guys can -- first of all, forget the bit piece on CMOS imaging; it's not even relevant.

It's got nothing to do with any of it.

So if you look at the allocation that relates to imaging, you've got to totally exclude that.

It's not even a relevant measurement.

With respect to NAND versus DRAM, hey, look, you guys can do the math.

Virginia is capable -- we've already said is capable of about 7000 per week.

Lehigh is capable of about 11,000 per week.

And then we've got the Boise ramp from -- Kipp already mentioned we're doing about 5000 now; we'll take that up to probably 10, 11, 12,000 per week on 8 inch.

And it really all has to do with timing.

We've already said that we want to try to start ramping Virginia sometime in mid-'06, and Lehigh at end of '06.

So take your historical perspective on how fabs ramp, plug that in, and you ought to have a pretty good number.

So that comes out to be about, all the numbers together, about 22,000 in addition to U.S.

You're probably running somewhere out of the 22, one-fourth of by end of next year.

So 70,000 -- out of the 70,000, say about one-sixth or so would be -- is that the right calculation?

Am I in the ballpark?

If you're going to ask me to comment on the math you're doing in your head, I'm going to have to pass.

All right, thanks.

Excellent quarter.

Thanks a lot.

I appreciate it.

Happy holidays.

Gus Richard, First Albany.

Could you talk a little bit about the drivers of CMOS image sensors in the coming quarter -- is it market penetration, market share gains, or just really strong cellphone demand?

All of the above.

For us, really, as I mentioned in the near term to date and in the near future, the primary driver, again, is going to be camera-enabled cellphones.

And my perspective is that we are the beneficiary, certainly, of a growing penetration rate in terms of cameras, cameras being the numerator, total cellphones being the denominator.

We also have put ourselves in the position where we're the supplier of choice, just based on the imaging quality that we are able to deliver with our sensors.

So we are benefiting from market growth.

We're benefiting from market share growth.

And if I look out, say, a couple quarters to the, say, four to five to six quarters down the road, we will start to see some significant growth as well from automotive applications, as well as some further penetration in digital still cameras and in PC camera area.

And then any color on the split between DDR and DDR 2 in the quarter?

In the quarter we just completed, it was a little less than 50-50, DDR being less than -- DDR being greater than 50, DDR being less than 50% of the commodity piece.

In the current quarter, it's probably going to be about split 50-50.

Final question.

In the strange world of commodity memory, giving you are allocating capacity to IMFT, in the future, if there was a shortage of DRAM, could you revert that capacity to DRAM?

Or is that just a ridiculous thought?

We cannot -- well, put it this way.

It's not that it couldn't happen; it's that we have a partner manufacturing NAND.

And the NAND has to be manufactured within a partner.

So, to the extent that we are going to do something with the capacity, either increase it or decrease it, we have to have a discussion with our partner.

So, in a sense, it limits your flexibility from reallocating capacity going forward from DRAM to NAND and back and forth?

When you say limit, only to the extent that our partner believes that we should do something different with it.

In other words, there's somewhat of a self-correcting mechanism.

Again, not too belabor this, but if the margin on NAND is much better than the margin on DRAM, of course we would want to allocate it in the direction of NAND.

And if that were to be the case, the probability is our partner would be fine with that.

If the margin on NAND is worse than the margin on DRAM, it depends on how worse it is before our partners are going to reallocate it.

But clearly, if it wasn't doing very well at all because the market is oversupplied, they are unlikely to have too big a problem with the reallocation.

Bill Dezellem, Titan Capital Management.

We had a group of questions.

First of all, Mike, you had referenced camera phones with two cameras.

Would you please walk us through the applications where having two cameras on the same phone is advantageous?

And then, relative to the JV, the IM Flash JV, what percentage of the production needs to be dedicated to Apple versus the availability that you have to address the large opportunity that you are anticipating developing here later this year in the cellphone market?

On the first one, the dual camera phones, the primary application would be videoconferencing.

So we have what would call an in camera, which is typically a low resolution VGA camera, which would be on the inside of the phone, basically looking at you or taking a picture of the user for videoconferencing purposes.

And you have an out camera, which would be for taking photographs by the end user.

So penetration rate of dual camera phones today is not that great.

But again, as infrastructure improvements are made to enable more seamless videoconferencing, we expect to see that increase.

But that is one of the big drivers of the extended life of VGA cameras, is the dual camera phones.

On the --

Before we shift to next question, you referenced the infrastructure out there.

With the current technology, with the service providers, is this something that can be fluid?

Or are there upgrades to the existing infrastructure that really needs to be made?

It depends where geographically.

Probably the most advanced infrastructure today for mobile communications would be in Japan, and videoconferencing over the mobile network there is a reality today.

In Europe, I believe it is as well.

In the U.S., again, this is beyond my area of expertise, but it's probably quite some ways off into the future.

That's helpful.

And then the IM Flash JV?

This is Steve.

On the question regarding -- I think it was specifically with respect to the supply to Apple that we started and we had a long-term agreement with them on that, obviously, I'm not going to disclose a lot of specificity, because the confidentiality agreement is both with our customer and with our partner.

But I think, suffice it to say that the percentage of exposure that we have to that agreement is really controllable by us -- in other words, I mean the partnership, because its quantity defined.

So if we want to produce more product, then the percentage goes down; if we produce less product, then the percentage goes up.

It's completely within our control as to how we want to -- how much of our output that we want to be exposed to that.

So I think that's the best way to think about it.

That is helpful.

Thank you.

And then one additional question, and I think this was somewhat addressed earlier, but I'd like to try again to see if I can gain a little more clarification, please.

Relative to the inventory decline, given that the pricing in commodity DRAM was down, oftentimes you don't see inventory declining when pricing is also declining.

Would you walk us through again where, within which products you actually saw the decline in inventory, please?

I think we pretty much -- well, we ended the quarter with virtually no inventory on image sensors, as you could probably imagine, based on the way we are speaking about it.

So there was not much of a change there.

I think that, if I remember back a quarter, it probably was the case on the specialty DRAM as well.

So the inventory we carried was primarily in the PC DRAM area, and demand was strong enough to absorb it, albeit there was a tremendous amount of price pressure.

Had we projected that pricing was going to be increasing in the current period, I probably would have chosen to hold some more inventory.

But we weren't projecting that, still not really projecting that.

So as the market had strength enough to absorb it, we just sold at the market prices and it ended up being a drawdown on inventory as opposed to either building or keeping it flat.

But if we did hear your earlier comments in the call, there certainly appears to be the possibility that some of the manufacturers that are shifting from DRAM to NAND, that that may start taking place and actually impacting the market here favorably in this seasonally slow period.

I'm not going to predict that, but I certainly would hope that would be the case.

I think we have time for one more question, and then we'll wrap it up.

Ben Lynch, Deutsche Bank.

On the NAND side, this agreement that you have with Apple -- I know, the way Apple operates, you can't really say too much about that.

Could you just help us to think a little bit -- is this a sort of a cost-plus thing, or how you are supplying to them, or driven by market ASPs?

And basically, do you need to be cost competitive to their current suppliers to make comparable margins or not?

And I do have a follow-up, please.

Well, as you suspected, I'm not going to comment on pricing to a particular customer; that wouldn't be, probably, acceptable for them or us.

But having said that, on the cost side, our expectation is to be equal or better than anybody else on the cost side.

It's got nothing to do with an Apple agreement.

And I think the way to phrase it is we plan on making money on the product we sell to Apple, and we plan on making money on the product we sell to other customers.

So we expect to be as competitive as anyone in terms of cost.

And I'll try one for you, Bill.

I know you don't want to give specifics on gross margins by product area.

Could you maybe give us a rough estimate on the spread in gross margins between -- of those four major areas, the highest and the lowest?

No, we won't give you specifics on the spread, although Kipp indicated that the CMOS image sensors, the specialty DRAM and the NAND were running pretty close together this last quarter at a very attractive rate.

And there's a significant spread between that and the commodity DRAM.

Now, with that, we'd like to thank everyone for participating on the call today.

If you will please bear with me, I need to repeat the Safe Harbor protection language.

During the course of this call, we may have made forward-looking statements regarding the Company and the industry.

These particular forward-looking statements and all other statements that may have been made on this call that are not historical facts are subject to a number of risks and uncertainties, and actual results may differ materially.

For information on the important factors that may cause actual results to differ materially, please refer to our filings with the SEC, including the Company's most recent 10-Q and 10-K.

Thank you for joining us.

Thank you.

This does conclude today's conference.

You may disconnect all lines at this time and have a great day.