萊迪思半導體 (LSCC) 2003 Q3 法說會逐字稿

Good morning, everyone, and welcome to today's conference call.

Copies of the Lattice Semiconductor third-quarter ending September 30, 2003, earnings press release may be obtained from the company's website, which is www.latticesemi.com.

This call is being recorded, and it is being broadcast live over the Internet by CCBN.

A live broadcast and replay of the call will be available on the Lattice investor relations website, www.latticesemi.com.

At this time I'd like to turn the call over to the Senior Vice President and Chief Financial Officer, Mr. Steve Skaggs.

Please go ahead, sir.

Thank you and good morning, everyone.

Joining me today on the call are Cyrus Tsui, our CEO;

Steve Laub, our President;

Rodney Sloss, our Vice President of Finance; and Jan Johannessen, our future Chief Financial Officer.

Before we begin I would like to read the Safe Harbor statement; then I will provide a financial review and our outlook.

Steve Laub will then provide a business review; after which we will hold a question-and-answer session.

This conference call may contain forward-looking statements with the meaning of the federal securities laws, including statements about our future quarterly financial results, revenues, customers, product offerings, and our ability to compete.

Investors are cautioned that actually events and results could differ materially from these statements as a result of a number factors, including general economic conditions, overall semiconductor market conditions, market acceptance and demand for our new products, our dependencies on our silicon wafer suppliers, the impact of competitive products and pricing, technological and product development risks.

Please refer to our current filings with the SEC for further descriptions of these risk factors.

I will now review our statement of operations for the third calendar quarter of 2003.

Revenue for the quarter was $51.0 million, down 12 percent from last quarter.

During third quarter, CPLD product revenue accounted for $34.8 million or 68 percent of total revenue, and declined 14 percent sequentially.

FPGA products contributed $9.2 million or 18 percent of revenue, and grew 3 percent sequentially.

Low density simple PLD products accounted for $7.0 million, or 14 percent of revenue, and declined 22 percent sequentially.

For the quarter the Americas made up 37 percent of revenue;

Europe, 23 percent of revenue; and Asia, 40 percent of revenue.

In calculating the geographic mix of our business this quarter, we noted a mechanical error with regard to classification of revenue from certain transfer business.

This error also impacts the geographic revenue mix we previously reported for Q1 and Q2 of 2003.

I'd like now to read the correct numbers for Q1 and Q2 of 2003.

The correct numbers for Q2 are North America, 41 percent of revenue;

Europe, 26 percent of revenue; and Asia, 33 percent of revenue.

For Q1 2003, the correct numbers are North America, 38 percent of revenue;

Europe, 30 percent;

Asia, 32 percent of revenue.

We apologize for any confusion this error may have caused.

Revenue by end market for the quarter was as follows.

Communications, 49 percent of revenue; computing, 22 percent of revenue; and industrial other, 29 percent.

Revenue mix by channel for the quarter was as follows.

Direct, 56 percent of revenue; and distribution, 44 percent.

This quarter for the first time, we have begun disclosing revenue by product classification.

We believe this information will provide better visibility into the growth of our new products.

For the September quarter, new products accounted for 12 percent of total revenue, while mainstream products and mature products each accounted for 44 percent of revenue.

Proceeding with the rest of the statement of operations, gross margin for the quarter was 59.5 percent, down slightly from the 60.0 percent reported last quarter.

The decline in gross margin percentage is due to lower fixed cost coverage on a lower quarterly revenue.

Quarterly R&D expense was $21.2 million, and SG&A expense was $12.1 million for the quarter.

In aggregate, operating expenses declined to $1.0 million when compared to the numbers recorded in the June quarter.

Other expense for the March quarter was $3.6 million.

During the quarter we completed the early call of the $172 million balance of our 4 3/4 percent convertible Notes.

This action resulted in a nonrecurring charge of $5.7 million, which was included in that other expense figure that I gave you.

In addition, during the quarter we repurchased $16 million of our zero coupon convertible Notes at a discount.

The tax provision for the September quarter was a credit of $3.3 million, due to the release of tax reserves.

Intangible asset amortization was $18.7 million, and the September quarter GAAP net loss was $21.9 million or 20 cents per share.

On a non-GAAP basis, the loss was $3.2 million or 3 cents per share.

The non-GAAP presentation excludes the impact of the intangible asset amortization; however it includes the previously mentioned $5.7 million nonrecurring charge for the early call of the 4 3/4 percent convertible Notes.

Turning now to the balance sheet, cash and short-term investments at the end of the September quarter decreased by $191.4 million, to approximately $279 million.

During the quarter we used $14.2 million in cash to repurchase certain outstanding zero coupon convertible notes; and also used $177.0 million in cash to call our old 4 3/4 percent convertible Notes.

We also made a final interest payment of $1.8 million in the quarter related to the older Notes that are now called, and spent $2.3 million on routing capital expenditures.

Netting together these items, cash flow from operations was a positive $3.9 million during the September quarter.

Depreciation for the quarter was $4.7 million; and inventory declined slightly to $47.9 million from the $48.3 million reported last quarter.

I would like now to briefly discuss our financial outlook for the December 2003 quarter.

We enter the fourth quarter with continued limited visibility.

Our best estimate at present is for the December quarter revenue to be approximately flat.

Given our lack of visibility, we will once again limit our guidance to the next calendar quarter.

For the rest of the P&L, we currently have the following expectations for the December 2003 quarter.

We expect gross margin as a percentage of revenue to be flat.

We expect operating expenses to be flat.

We expect intangible asset amortization to be approximately $18.7 million.

We also expect approximately 400 to $500,000 in other income.

We plan to report no tax expense or credit.

Finally, we expect the share count to be relatively flat.

With that, I would like now to turn the call over to Steve Laub for his comments.

Steve?

Thank you, Steve.

Good morning, everyone.

Steve has covered the results for Q3.

I would like to highlight two areas of strength in an otherwise soft quarter.

Our FPGA business grew by 3 percent sequentially and has now grown by 22 percent year-over-year.

In addition, our new products generate over 40 percent sequential growth, now representing approximately 12 percent of total revenues.

As stated earlier, in addition to our traditional product categorizations, Lattice has this quarter, for the first time, begun to disclose revenue by a new product classification, which is new products, mainstream products, and mature products.

The products within each classification are as follows.

New products are comprised of both FPGA and CPLD products based on 0.18 micron or more advanced process technologies, and also includes our mixed-signal power manager products.

Mainstream products generally consist of products based on 0.25 micron to 0.3 micron technologies for FPGA products, and 0.25 micron to 0.5 micron for CPLD products.

These include our ORCA 3 FPGAs and 3.3 volt CPLD products.

Mature products are generally based on earlier-generation process technologies, and include our ORCA 2 FPGAs as well as all 5 volt CPLD products as well as SPLD products.

As mentioned, we introduced this new product classification, as we believe it provides the investment community better insight into the dynamics of our productlines, especially with respect to customer acceptance and the revenue ramp-up of our new products.

Given this product classification, I will now share with you our performance by product class at the beginning of 2003 and for each quarter of 2003.

New products have climbed from approximately 3 percent of total revenues in both Q4 2002 and for the first quarter of 2003, to 8 percent in the second quarter of 2003, and now 12 percent in the third quarter of 2003.

Mainstream products have gone from 38 percent of total revenues in Q4 of 2002, to 36 percent in the first quarter of 2003, 40 percent in the second quarter of 2003, and 44 percent this past quarter.

While mature products have declined from approximately 60 percent in both Q4 of 2002 and the first quarter of 2003, to 52 percent in the second quarter of 2003, and 44 percent this past quarter.

As shown, new products have begun to generate substantial revenues and growth, again over 40 percent growth this past quarter alone; while mainstream product revenues have been generally flat and mature products have been experiencing a substantial decline.

As is clear, Lattice's results this past quarter have been heavily influenced by the significant decline of our mature products, specifically 5 volt CPLDs and SPLDs, which were primarily driven by the part transitions of two European customers.

We mentioned last quarter that the full impact of these customers' transitions were likely to be felt in Q3; and they were.

Overall, design-in activity this past quarter was strong, and up from the strong performance we recorded in the second quarter.

Also, our mix of designs continues to shift in favor of new products.

Based on our newly introduced product classification, the design-in of new products as a percent of total design-ins has climbed from 17 percent, as of the end of Q4 2002, to 50 percent this past quarter.

On a sequential basis the design-in activity from new products by approximately 20 percent.

Within our new products category, the design-ins of our FPGA and FPSC products grew by 85 percent sequentially and set another record high.

Based on the design-in success we are experiencing, we expect that in the FPGA marketplace our FPGA revenues will continue to increase on an annual basis.

However, as we stated in the past, since the business is still a mix of production and prototyping, it is nonlinear; and therefore the business is expected to grow annually.

It will not do so each and every quarter.

Sales by end market this past quarter showed a decrease in our communications sales as a percent of total sales, from 52 percent in the second quarter to 49 percent this past quarter.

Areas of relative strength included wireless, primarily 2.5G equipment; enterprise networking;

ADSL (ph) and consumer-based access equipment. 3-G is beginning to show some growth, while voice continues to be quite weak.

In computational-based revenues, as a percent of total revenues, they increased from 21 percent to 22 percent this past quarter.

This is primarily due to the growth in mass storage applications. (Indiscernible) change in our sales by end customer, with a proportional increase in our sales to industrial/other customers, which as a percent of total revenues increased from 27 percent in the second quarter to 29 percent this past quarter.

From a geographic standpoint, North America declined this past quarter to 37 percent of total sales from 41 percent in Q2.

European sales declined as a percent of total sales, from 26 to 23 percent, while Asia grew from 33 percent of total sales to 40 percent.

As you are aware, much of the growth in Asia is due to transfer of business from North America and European-based OEMs to project manufacturers in Asia.

And the growth is also due to indigenous companies based in Asia.

As mentioned, our outlook for Q4 is to be approximately flat with Q3 due to the following factors.

First, the backlog entering Q4 is basically flat with the backlog at the same time entering Q3.

As I mentioned our visibility continues to be limited.

While we expect continued strong growth in new products, these are likely to be offset by some decline in mature products.

Also on a seasonal basis, the numerous holidays present in Q4 provide more uncertain regarding the level of turns business we are likely to realize.

Therefore we believe flat is the most likely outcome.

Last week, Lattice announced that I will be leaving at the end of November and that Steve Skaggs will be assuming the position of President.

The basis for my decision is personal and family related, for reasons I will not pursue in more detail here.

However, as the decision to leave Lattice became necessary, Cyrus and I chose a time when we could minimize the impact to the company.

Despite short-term business conditions, the company is in good shape.

It is making excellent progress in growing its new products, the results of which I expect we will see even greater impact this coming year.

The management team is seasoned, stable, and experienced.

I'm very confident regarding Lattice's future prospects.

I will now open the session to questions.

(OPERATOR INSTRUCTIONS) Chris Danely of J.P. Morgan.

Could you just talk about the end-market expectations for Q4?

Also, it seems like all three of the big PLD guys have talked about a lot of strength in storage.

Can you give us a little color as to what is happening over there?

From an end-market standpoint, I think our expectation for Q4 is pretty similar with that which we saw this past quarter.

From a standpoint in the communications area, which is our strongest, -- excuse me, our biggest end-market, enterprise networking continues to demonstrate some strength.

In the base station area, 2.5G, CDMA 2000 are showing some strength.

As I mentioned 3-G began to show some stirrings this past quarter as well.

Voice was weak, and we can expect that will continue to be weak this quarter and into next year.

From a standpoint of storage, what we are seeing is in the large mass storage providers, that area is where we are seeing some growth.

And the expectation is they will continue to grow into this quarter.

So you expect most of the growth, or the best growth, to come out of the computing segment for Q4?

Is really hard to just, within the beginning of the quarter, to say this is where the biggest growth may be.

Because right now, there's a lot of things that drive things up and at the same time drive some things down.

We are seeing, as we saw this past quarter, slight growth in the computational segment.

Sense is some of that is going to continue.

But at the same time, you don't know what is happening with all the different customers in that area.

Specifically in the storage area.

I would think that perhaps that are may grow; it is expected to grow this quarter.

But the server area itself is a little bit more difficult to predict.

Thanks.

Mark Edelstone from Morgan Stanley.

I wonder on the inventory side that you guys have on your balance sheet, do you have the same breakout by product category?

So, new, mainstream, and mature?

No, we don't have that breakout.

Two other questions on the product classification that you have there.

Obviously mature has been under pretty good pressures, and I would think that that would continue.

But do you have a sense, based on some of the products that you are shipping into now, what the level of just stabilization would be?

I would guess there would be some level where those revenues would just kind of hold up; whether it would be $15 million a quarter or $20 million a quarter; some level where they will be relatively stagnant for a while.

Where do you think that is?

And when do you think we might get there, if that is possible?

With respect to the mature products, I think anybody who is predicting at what level a number stabilizes has been proven wrong the last few years.

But one thing that we have some comfort in is that we are looking at the backlog by these products.

What we are seeing is more stabilization with respect to the backlog on the mature products this quarter, as compared to the prior quarter, or the prior two quarters in fact.

So in that respect, we don't expect the drop off in the mature products area this quarter, as we have experienced certainly this past quarter.

Both because, one, we did, as we said earlier, we indicated to you guys that we were seeing some specific customer issues, transition issues, which impacted those products.

And we also did see the backlog dropping off consistent with that.

You don't see that same back off this quarter.

So that does give us some pause, that perhaps the worst is behind us on that.

Last question.

Obviously you have got a pretty strong ramp going on of the new products.

If you were to look at your new products in total, and try to compare that against a comparable ramp rate that you have had in your history, what would it most closely align with ?

Would that be looking at like the ISP ramp that you'd seen?

Or was there something else that we can look at, for a comparable growth pattern?

That's a good question, Mark.

Unfortunately a lot of things that have happened the last two or three years have thrown off the traditional models of looking at things.

With respect to the right, clearly it's demonstrating for us a very successful new products ramp.

It is a quite strong ramp of new products.

Consistent with that we've seen from, I would say, our more successful products in the past.

We have not a comparison of saying, how does this compare to the ISP ramp that we experienced in the mid to late '90s, which is that we are (indiscernible) to do this quarter.

This classification is new for us.

It is something that we have not done in the past.

So I think a lot of the kind of barometers of how you measure that and compare, a lot of that we will be doing this quarter.

We did this specifically to provide more insight for you guys.

Because a lot of people always ask questions, which we understand, saying design activity is great, but we want to see the revenue ramp associated with that.

So we want to be able to provide the two together.

That is why we did that.

Thanks a lot; that is actually quite helpful.

Thank you.

(OPERATOR INSTRUCTIONS) Sumit Dhanda of Banc of America Securities.

A couple of questions.

First off, on the CPLD side of the business, it seems like it was down pretty significantly.

Other than the two customers that you talked about, where there any other dynamics we should be aware of?

Xilinx was indicating on their call that pricing pressure within that segment has been very intense.

If you could comment on that.

With respect to CPLD pricing, we have seen some specific activity within CPLDs, primarily as it relates to competitors' actions with respect to written-down inventory.

When you put that aside, the pricing behaviors are pretty consistent with what you see historically.

But that typically has had a greater impact.

As that inventory gets utilized and is no longer there, we expect that those price pressures should subside.

But that is where we have seen the price pressure, specifically where written-down inventory is being utilized.

Also on the FPGA side of the business, it seems like if I go back four quarters to the December quarter, you were averaging about 9 million or so; and you are still at that level.

With is it going to take for Lattice to break the 10 million mark here?

It seems to be a real hurdle the last couple of quarter.

I think there is nothing magical about 9 million or 10 million.

It is really just a number that it adds up to.

We are experiencing growth in the FPGAs.

As I mentioned, it has grown over 20 percent over the past year.

We expect that the FPGA business will grow over the next year.

So with that, I would expect that you'll see in 2004 a growth that should put us over the 10 million mark.

But again, this is a business that is growing; it is a business that will be nonlinear in its growth.

But again you should expect that it will experience and show that type of growth for us.

In terms of where you finished on the September quarter, initially you had given guidance for 51 5o 55 million; and you finished below the range.

Other than the customer transitions, what were some of the other factors which arguably caused a slippage towards the end of the quarter?

Primarily the seasonality of the summer quarter; it was probably a little bit softer than we expected going into the quarter.

But that is primarily what it was.

I think actually that is something that other people have indicated as well, that the seasonal softness of this summer quarter this year was a little greater than they experienced.

Thank you.

David Wu, with Wedbush Morgan.

Can you clarify a couple things for me?

On the new products, where you talk about the mixture of CPLD and FPGA, what is the mix in the new category between those two classes of products?

I have got a couple more as well.

Do you mean what is percent?

What amount is the CPLD versus what amount of that is FPGA?

In the new category.

Give me a moment here.

I would say, roughly, what you would have is that the CPLD is probably about three quarters; and FPGA is about a quarter.

Also on this whole category called communications, should I think about there is three bucket of waters, base station, voice, and enterprise networking, about equal weighting in that category?

What I would say is probably not with respect to that.

What they are is, they are the major -- we think about this business as these are the major categories within that; and they have different behavior with respect to that.

But they are not necessarily the same size with respect to our sales into that.

I assume your biggest piece at this point would be enterprise and base station, relative to voice, right?

That is correct.

Last question I have is, with all these design-in activities, in the past your leadtimes on the class of FPGAS you guys are dealing with is 12 to 18 months from design win to production.

Is that still the right kind of order of magnitude we are looking at?

With respect to FPGAs, what we have been experiencing is closer to an 18 month plus, with respect to real ramp-up of production.

CPLDs are probably closer to 12 to 18 months.

FPGAs are really 18 months, even closer to 18 to 20, 22 months.

It actually takes longer than we expect, I think primarily because the FPGAs that we currently focus on are what is called our FPSCs.

So field programmable system chips, which have a lot of complexity and a lot of functionality embedded within them.

So they actually begin and those designs begin much earlier in the entire design cycle than a traditional PLD does.

I see.

So in terms of the ramp, that should hit your P&L probably, what, first half of this next year?

We certainly expect to be seeing some production ramp of those designs, since we acquired the business, yes, in the first half of 2004.

Thank you.

Robert Toomey of RBC Dain Rauscher.

Just a follow-up on that series of questions.

Can you talk about, has there been any change in the application for CPLDs that would account for some of the softnesses?

Can you comment on that?

I would say the application for CPLDs haven't really fundamentally changed.

In fact, in some respects there was some expansion.

This particular quarter that we just finished was a relatively soft quarter for the CPLD market.

But if you go back, I forget which quarter it was, two or three quarters, there was a substantial growth in CPLDS; primarily being driven by a lot of zero-power and portable and battery type of applications, which historically CPLDS did not participate in at all.

Those are the kind of applications that our zero power CPLDs are focused on.

So there has been an expansion in some applications for CPLDs.

With respect to some of the other applications for CPLDs, I would say there has not been a big change.

Historically it was much more computational than it is today.

It still is a communications focused type of product, from the standpoint of that is still its largest end-market.

But I would not say there's any applications where CPLDs don't participate in that they did participate in, say, a year or two ago.

So you don't see any diminishment in the potential end-market applications per se?

That is correct.

There's been some talk about supplier OEM consolidation in your industry.

Can you comment on that?

Do you think there's been any impact on Lattice with respect to this?

Supplier OEM, specifically you are talking about?

The large OEM customers consolidating their purchasing from CPLD companies.

That is something, I think, that our competitors would like to espouse to the customer base.

But with respect to the activity that we are seeing on that, in fact, I would say that our penetration and our relationships with the large OEMs is actually a broader relationship and a deeper relationship today than it was two years ago.

There is always some talk, whenever there's a downturn in the business, about supplier consolidation.

At the same time, whenever a business begins to turn up and availability begins to get a little bit tighter, there's always talk about supplier expansion.

So these things go with the supercality (ph) of the business.

I think you will be hearing much more about people expanding their suppliers to make sure they maintain supply, especially in the 2004 time frame, to the extent that business does in fact continue -- for the semiconductor industry to grow.

Great, thanks very much.

(OPERATOR INSTRUCTIONS) Jim Colgan (ph) of Frontier Capital.

Could you talk a little bit about the mainstream products?

I know they are going to be flat this quarter, but what has been going on in the last three quarters?

And what do you see for the future?

And where are those more targeted to?

What kind of end applications?

The mainstream products have been generally flat, or actually, I think, even slightly up if you go back. (indiscernible) actually up slightly; pretty much flat for the last three quarters.

With the applications in those, the applications for those products are not fundamentally different from the applications for our new products.

They are more established systems.

With respect to the CPLD business, it is 3.3 volt, which is still a very popular power supply voltage level for those products.

So in fact I think in that area you may see, to the extent there is a resumption in the overall market growth, some growth in the mainstream products.

Especially as it reflects CPLD.

With respect to the mainstream products and the FPGA area, those products are some earlier generation for us.

Those being primarily our ORCA 3 products.

Those actually are much less likely to experience much growth in the future.

The growth in the FPGAs will really be in the new product area.

So that's one thing that you will see.

Because the ORCA 3 products are really focused on or really have been designed into applications which are more steady into manufacturing today, and production-related systems.

But they are not really going to that many new designs, as compare to what the new product FPGAs are.

Industrial and other, could you break that a little bit further in terms of the end applications there?

The end applications, industrial and other?

For industrial you have industrial control.

You have of factory automation type of systems.

You have medical systems.

We also have within this industrial/other category automotive systems.

Telematics.

Those type of applications.

You also are having consumer applications; both the battery-powered and portable systems.

Those would be, for example, digital cameras.

They could be within an application some type of special sort of patient-monitoring equipment, which would be in the medical area.

So there's a lot of applications, a very wide spread of applications, using PLDs within the marketplace.

Does that help?

Yes.

What are the gross margins on the new products?

Are they similar to what you are seeing today?

That is correct.

We have not broken it out.

It has been pretty much similar.

Is there any indication from your voice customers -- they seem to be the weakest part -- if inventories are low now, and they have got to rebuild?

Or is there any signs of any pick up there?

We are not seeing much pick up in that particular part of the marketplace right now.

Certainly we hope that there is a growth in that area.

But I think everyone is seeing relative softness in that area.

We are seeing what pretty much everybody else is seeing as well.

Any plans for the remaining bonds outstanding?

We will continue to be opportunistic concerning repurchases, depending on the value of the bonds.

Thank you.

David Duley of Wells Fargo Securities.

Could you talk a little bit more about your European transitions?

I am assuming they are moving out of 5 volt CPLD products and perhaps moving to lower voltage products.

Could you tell us the timing of that?

When we would expect that kind of bucket of revenue to be back at normal levels?

Dave, you're correct.

The systems that they transitioned out of were using 5 volt based products.

The new ones are using 3.3 volt potentially.

That's primarily what they are using. (indiscernible) any 1.8 voltage.

Primarily 3.3 volt based products.

The expectation in fact was that the transition was not going to be quite as strong an impact initially as it turned out to be.

The new systems that we are designed into have not had the same kind of success.

Now, partly, the reason that we experienced such an abrupt drop was not only, one, was the new system not ramping up as fast, but they are using up their inventory of our existing 5 volt products that they had purchased in the past, as they end-of-life their system.

So you get kind of a double whammy effect on that.

We do expect to see some recovery from those customers in the first of next year.

Actually documenting how much at this point is very difficult to do, so I won't do so.

But we do have indications that they will be resuming some ordering of product in the 3.3 Volt area in the first quarter of next year.

In the mainstream and new product areas, do you have plans to migrate those products to smaller geometries?

What kind of cost advantages could you get from that migration path?

With respect to our decisions about which products we migrate to smaller geometries, we do that based on several factors.

One being the overall popularity of that particular product.

Obviously those that have the highest volume potentially give you the greatest opportunity for cost reduction.

We look at that.

We look at the overall value and impact of the cost reductions.

So some products will be taken for cost reductions; but we won't necessarily say okay, all the new or all the mainstream.

We look at, within those productlines, which ones make the most sense for us, based on our investments into both future demand and the opportunality for cost reduction.

Thanks

Neil Jacobs (ph) of Baudry (ph) Capital Management.

Assuming that your goal is still a 20 percent operating margin, you are about 60 percent.

Well, you would need to do on the top line 60 percent more in revenue with your current cost structure to get to that 20 percent.

At what point do you began to reassess the cost structure you have?

It is a fair question you are asking, from the standpoint of given a certain revenue level and so forth of the company and the investments we're making.

Now as you know, one of the key things that that is going to depend on is really a perception of two different areas; one being, do we think that the business tone is going to improve or the business tone is going to stay the same.

And while our outlook is flat, there is some indication that the business tone is beginning to improve.

So, we do believe that next year will be a better year than the last few have been.

So that gives us indication that perhaps we will see some growth in the top-line, which will help relieve or alleviate some of the issues with respect to margins.

The other thing is that, one of the key reasons we have restructured some of those expenses is that we're making very substantial investments in the FPGA area.

The FPGA opportunity for us is an enormous one.

We only recently entered this marketplace.

We have critical mass in this company to be able to compete effectively in that marketplace.

And we think that it is in the best interest of ourselves and of our shareholders that we continue to make the investments necessary to assure a very successful position in that marketplace.

And that to cut, at a time when we are in the midst of making those investments, and in the midst of seeing success from our investments, would be very short-term guided.

So that is our position today, and it is something that we do reassess, being thoughtful businessmen.

But that is our position today.

Thank you.

Robert Toomey of RBC Dain Rauscher.

I had a follow-up.

I wondered if you could comment, Steve, on ultimately -- you said a second ago that you see the tone of business improving a little bit.

What would you really see driving an upturn in the business substantially?

What kind of business conditions would you see to make that occur.

If you could comment on that.

Fundamentally, the biggest end-market for PLD, and that which we all talked about in very positive terms when you back to the late '90s and the year 2000, was the communications marketplace.

Which at the time represented nearly 70 percent of the total consumption of PLDs.

It is the biggest end-market today.

It has been a relatively soft end-market.

If that end-market begins to grow, grow in a substantial way, I think you will see the PLD marketplace, where all the PLD players, at least the top three, have a very positive response.

But that would have the single greatest impact, just seeing growth in the fundamentals of the communications marketplace.

The new products you have developed in the FPGA area, are most of those targeted at communications?

A number of our new products in the FPGA are.

Specifically, our FPSC products, for example.

Those are targeted at communications.

Our XPGA products actually can serve many different application areas and are tuned for many different application areas within PLD.

But the FPSC is, yes.

And that is one of our new product areas.

You still are carrying a lot of cash.

Would you be using that mostly for, again, opportunistically buying back debt, primarily?

Or can you comment on the cash?

Steve indicated that buying back debt is something we would do on an opportunistic basis.

But fundamentally, having cash is a good thing to have, especially during the kind of business conditions that the industry has been in for the last few years.

So it's more that we think it is prudent to have the cash that we have.

We use it in a very thoughtful way.

To the extent that we buy it back, we will buy back those bonds because it makes good business sense to do so.

And finally, tax rate for modeling purposes?

Zero.

Thank you.

Sumit Dhanda of Banc of America Securities.

Quick follow-up.

On the turns number, what was it last quarter?

And what are the expectations for the December quarter?

About 60 for both quarters.

Thanks.

Gentlemen, at this time there are no other questions in the queue.

Thank you.

Thank you.

That concludes today's conference call.

We thank you for your participation.

You may disconnect at this time.