Cypress Semiconductor Q3 2007 Earnings Call Transcript

Cypress Semiconductor Corporation (CY)

Q3 2007 Earnings Call

October 18, 2007 11:30 am ET

Executives

T. J. Rodgers - President, Chief Executive Officer, Director

Brad W. Buss - Chief Financial Officer, Executive Vice President - Finance and Administration

Christopher A. Seams - Executive Vice President - Sales and Marketing and Operations

Norm Taffe - Executive Vice President, Consumer and Computation Division

Ahmad Chatila - Executive Vice President, Memory and Imaging Division

Dinesh Ramanathan - Executive Vice President, Data Communications Division

Analysts

John Barton - Cowen & Co

Chris Danely - JP Morgan

Adam Benjamin - Jefferies & Company

Tim Luke - Lehman Brothers

Sandy Harrison - Signal Hill

John Pitzer - Credit Suisse

Srini Pajjuri - Merrill Lynch

Glen Yeung - Citigroup

Douglas Freedman - Amtech Research

Presentation

Operator

Good morning and thank you for standing by, and welcome to Cypress Semiconductor’s third quarter earnings release conference call. (Operator Instructions) I would now like to turn the call over to Mr. T.J. Rodgers, President and CEO of Cypress Semiconductor. Thank you, sir, you may begin.

T. J. Rodgers

Good morning. We’re here to report the third quarter of 2007, in which we recorded record revenues. As usual, we’ll do finances by Brad Buss, our CFO; and then the market by Chris Seams, our VP of marketing and sales. I’ll add a few technical notes and then we’ll go to questions. Brad.

Brad W. Buss

Good morning. Thanks for attending. As usual, I just want to do the Safe Harbor and forward-looking statement disclaimer. Please make sure you take a look at our press release, as well as our SEC filings, for a detail of all of those. We also have a full reconciliation of all the non-GAAP to GAAP measures in the press release and on our website, so please feel free to check them out.

I’ll take us through Q3 and then I’ll take a quick look at guidance for Q4. As you can probably see for Q3, I was very excited as the CFO to see the results that we posted. We had consolidated GAAP revenue that hit an all-time high, as T.J. mentioned, at $450 million and exceeded our guidance as well as the Street First Call estimates quite significantly.

Revenue increased 21% sequentially, 55% on a year-over-year basis. SunPower continued to grow strongly and accounted for 52% of the consolidated revenue, which was a new record.

Our semiconductor revenues totaled $215 million, which is a strong 8% sequential increase, and it was actually 12% if you adjusted for the PSRAM divestiture in Q2. It did not repeat again in Q3, so a very strong sequential growth.

All divisions grew sequentially on a divestiture adjusted basis. BCD was the star. It surged 21% sequentially, driven by PSoC and USB, which both set new quarterly records due to really strong design wins ramping in cell phones and PCs, and obviously a very strong consumer seasonal build that we normally see in Q3.

BCD grew 8% sequentially, driven by West Bridge production ramps that were consistent with our expectations, and MID actually grew revenue in SRAM by 6% and offset a good portion of the revenue loss from the PSRAM divestiture that I talked about.

On a GAAP basis, we had net income of $30 million with diluted earnings per share of $0.18, versus last quarter’s $2.29, which you’ll remember was favorably impacted by the sale of the SunPower stock. On a year-over-year basis, that’s up significantly from the $0.06 that we posted in the third quarter of ’06.

Our non-GAAP net income was $45 million and resulted in diluted earnings per share of $0.26, and again this is the highest diluted EPS on a non-GAAP basis that we’ve seen since Q4 of 2000, so a very big increase. And again, that compares to $0.16 in the prior quarter, as well as $0.16 in the year-ago third quarter.

We’ve been talking pretty heavily about how we’ve expected to see a lot of the strategic actions start to flow through to our financials, and I think as you see, that happened quite nicely in Q3. The semi business, which we break out separately from SunPower in the press release, actually increased sequentially on a non-GAAP net income basis by 100% and contributed $0.18 of the $0.26 that reported, up from $0.09 in Q2. And I expect to see continuing operating leverage from the semi business during 2008 as their revenue and gross margins are expected to increase.

The consolidated non-GAAP gross margin was 33.2%. It was down from the previous quarter, basically just due to the mix of SunPower business which, as I mentioned, was 52% of the business versus 47% in Q2. Obviously it’s important that most of you do, you look at SunPower’s gross margins and ours separately. And again, we provide all of that information easily accessible in the press release for you.

The non-GAAP gross margins for our semiconductor business was 47.2%, up from 44.9% in the previous quarter, as we expected, due to favorable product mix and higher utilization in our fab.

Our direct margins continued to remain strong across all divisions and our corporate ASPs increased again sequentially, and that’s the seventh consecutive quarter that that’s happened.

Our non-GAAP consolidated operating expenses in dollars, which is R&D as well as SG&A, increased from the prior quarter, mainly due to SunPower but also to some slight increases in the semi business, just due to timing of certain expenses in the quarter.

Our total consolidate non-GAAP op-ex is at 23% of sales, our lowest level ever. Semiconductor R&D in absolute dollars continues to stay at historically low levels, even while we continue to invest heavily in design and software for future programmable solutions that will continue to drive longer term revenues and gross margin expansion.

Turning over to the balance sheet, it’s looking really nice. We had consolidated cash, equivalents and investments that totaled $1.4 billion, which is all-time record high. The semiconductor cash and short-term investment balances continued to grow and totaled $942 million -- again, our highest balances ever.

Operating cash flows from the semi business was a strong $52.3 million for Q3, due to our strong profits and tight working capital management.

Turning over to inventory, our net consolidated inventory was $196 million, which was down $6 million from the prior quarter and of that decrease, the semiconductor business was $5 million of that. So we managed inventory very well in the quarter.

We had stated that we planned to bring inventory down, which we did. We took it down in MID, partially due to sales but also partially due to managing our profiles, and we increased slightly DCD and PSoC, as we expected to support our PSoC and West Bridge growth ramp.

Inventory in the channel with our disti partners decreased and is down over 18% from levels in mid 2006, so that’s a positive indicator of low inventory in the supply chain, which I think is goodness. I expect semiconductor inventory dollars to be flat to slightly down in Q4.

Accounts receivable was in good shape again. It decreased as well as DSO did, and no issues there.

CapEx was $65 million for the quarter, of which $59 million was for SunPower, and we only spent $6 million in the semiconductor business and again, consistent with our discussions of flexible manufacturing and [inaudible], we have a very low CapEx spend this year and I expect to see that continue in the quarter.

Depreciation was broken out in the press release for you and it was basically $25 million in total, of which 19 was the semi business and the balance was SunPower.

With respect to our ownership of SunPower, our basic ownership was 57%. Fully diluted, it was 53, and it declined slightly from Q2 due to SunPower’s offering. Cypress did not sell any shares in SunPower during Q3. We currently control 90% of the voting rights due to our Class B structure. We continue to won 44.5 million Class B common shares and as of yesterday, they had a market cap of $4 billion.

We had a really nice stock price increase in Q2 and Q3 so far. The one negative of that is that goodness obviously impacts the share count due to the higher stock price. So the average stock price in Q3 for EPS was $25.76 and that was up from $21.50 in Q2, so it impacted the options, it impacted convert, which became slightly dilutive for the first time. And the convert, as you know here, the conversion price is $23.90, but unlike our old convert, there is no interest add back, okay? So people need to kind of remember that because this is cash settled and you use the treasury stock method. So don’t mess that up in your model.

And another note, under the accounting rules, the note and the related call spread must be treated separately for the EPS calculation. So for the call spread, neither the warrant nor the call option impacted EPS in Q3. However, the warrants will have an EPS impact once the quarterly average stock price is greater than 27, which we currently are at in this quarter.

Under the accounting rules, the call option portion is not included in diluted EPS because it is anti-dilutive. So as such, the accounting treatment for EPS actually paints a worse dilution impact than what will really happen upon settlement in 2009. So again, just to reiterate, upon the settlement, the net economic share dilution of Cypress will be zero, up to a stock price of 27 due to the call spread, and it will move up with every dollar of stock appreciation, which in my mind is not a bad thing to have. I kind of like seeing the stock up in the 30s.

So just to give an example, at a settlement in a stock price of $37, the net economic share dilution will only be 5.7 million shares, which is substantially less than the 13.7 million shares that would be required under EPS accounting rules at that same stock price before settlement.

I know it’s confusing. I wanted to spend a little time on it because I’ve have a few calls. And we’ve actually posted some worksheets to the website so you can look at the stock price you want to model, look down the column, boom, here’s some numbers and you can put it in your model. If you have any questions, please feel free to give me a call.

Guidance for Q4 -- we are pleased to report that our estimated performance for the second half of 2007 will be higher than the First Call estimates before we enter the call. So just to put that in perspective, the First Call estimate on a non-GAAP basis for the second half was for revenue of 838 consolidated and diluted EPS of $0.45. We now estimate the second half we’ll have revenue of 875 to 892 and the diluted EPS will move up to 52% to 54%. And again, that even includes a little higher share count than I think any of us had expected due to the stock price increase, so I am very pleased with that guidance.

So the revenue for Q4 is expected to be in the range of 425 to 442, which is a year-over-year increase of 48% to 54%, and it is equal to the First Call or up to $17 million higher than the current First Call estimates for Q4, due mostly to higher revenue in the semi business.

SunPower revenue will decrease, as they previously guided, due to the timing of certain large projects, and our semi business will be flat to up 3%, which I was very pleased to see because we did have a blowout Q3 in that business. So things are looking very good from that perspective.

And also, don’t forget that’s for the piece of the NFC business late in Q2 or Q3 that doesn’t repeat going forward, so please adjust your models.

Gross margins will increase for SunPower and are expected to be in the range of 24 to 25. We expect semiconductor gross margins to move up again and be in the range of 48 to 49, and we do continue to expect ASPs and customer direct margins to remain strong.

No real major changes to the tax rate. Q4 should be around 8 to 9 for the semi business, 11 to 12 for SunPower, and our basic share count -- this will be kind of a wildcard, depending on the stock price, but the basic share count should be around 158. But it could be as high as 177 to 185 on a fully diluted basis, again due to a lot of the impacts of the convert in the call spread, which I talked about, which obviously the economic impact on settlement will be vastly different.

CapEx for 2007 should be at $50 million or slightly less, which is again is a decrease of over $60 million from 2006 as we continue to enjoy the benefits of some of our new strategies and I expect CapEx to remain flat around that level going forward.

As I mentioned, we continue to own 44.5 million shares of SunPower and we currently have no lock-up agreements with respect to those shares. The prior lock-up agreement expired at the end of September.

Our consolidated fully diluted non-GAAP EPS is expected to be in the range of $0.26 to $0.28 cents and again, that’s currently higher than the First Call estimate of $0.24.

All the guidance is obviously subject to change due to a lot of things that you can take a look at in our risk factors. And I just want to wrap it up by saying I was really pleased with the performance that the Cypress team has turned in, as well as the SunPower team. We had some really good numbers and I think we are shaping up for a very strong 2008, assuming the economy holds up.

And I’ll turn it over to Chris now. Thanks a lot.

Christopher A. Seams

Thanks, Brad. Let me give some of the usual indices for the semiconductor business for the third quarter.

Our revenue splits by geography: Asia climbed to 55% of sales, driven by a large ship-to in the consumer end segment and PCs, PC peripherals and handsets also aided that. North America was 25% of sales and Europe and Japan were 11% and 9% respectively.

Again, we have no customers greater than 10% of sales. Units climbed in the quarter as our revenue grew. We were up to 166 million units. That’s up from 156 million in the previous quarter. As Brad noted, the pricing environment continued to be very stable for us. We had another rise in ASP to $1.29. That’s up from the second quarter at $1.26.

Book-to-bill was stable at 1.01 and our semiconductor backlog was slightly down at $229 million as we brought lead times slightly down in the quarter. And we enter the fourth quarter 78% booked, which is still a real strong booking number and the order patterns, booking patterns two weeks into the quarter appear to be normal so far.

Let me turn the call back to T.J. now for more details on the quarter.

T. J. Rodgers

Let me give a little bit more financial information first on the divisions. First, we have DCD and consumer computation, that’s where our programmable products, including PSoC, are. It did $103 million, up 20.5% quarter on quarter, still very strong growth, 49% gross margin.

We no longer report because we can’t come up with a GAAP number by division. We no longer are allowed to report, I should say, profitability of the divisions. I will give them verbally here. I think they are important and I think the SEC takes information away from investors in doing this. DCD had a 20% pretax profit in the quarter. Data com did $29.7 million. It’s only 6.6% of the company. They were up quarter on quarter 8.4%. They made 63% gross margin and put 16% in TBT.

Memory and imaging division did $79.6 million. That’s 17%, 18% of the company, and they dropped a little bit, 4.6% quarter on quarter. They delivered 41% gross margin, which is the lowest of our three divisions, but they are also organized the leanest and still produce 15% pretax profit.

Chris commented that our ASPs were up to $1.29, the seventh consecutive quarter of rising ASPs for us.

Business and technical comments, our PSoC customer base has now grown to 5,300 customers. That’s a lot for us. We are on our way. Our long-term goal is 40,000 customers.

Among our customers, global customers that have now engaged in PSoC are Hewlett-Packard, Cisco, Lucky Goldstar, Lenovo, Nintendo, and Pentax.

We shipped our 250 millionth PSoC in the quarter and a couple of the interesting designs we have that will be significant in volume are the Marion Global GPS system and Haier Group's washing machines. Haier is a Chinese company that makes appliances for the Chinese market, obviously high volume. And they used Capsense, our PSoC no mechanical button solution, waterproof, on their washers.

We introduced a first touch embedded design starter kit. It sounds like something that probably wouldn’t be interesting to investors but I invite you to e-mail me, [e-mail], and I will send you one and you will actually be able to plug it in and play with PSoC and see some of the features. You can plug it into the USB port of your computer. We sell it for $29.95 and it will show you the four different demos of what a single PSoC can do when you reprogram it.

We introduced PSoC Express III, that’s our latest generation of our software. This is software that allows designers in our customers’ companies to design a system by dragging and dropping block icons for chunks of the system and connecting them with lines and specifying the relationship by truth tables and state diagrams. In other words, it’s a totally visual design system that allows our customers to design a system without writing a line of code, hugely enhancing their productivity.

This software is five years in the making and we currently have over 50 people in the software. It is a huge differentiator for us.

As we know, the world is turning to more efficient solutions for lighting and that includes fluorescent bulbs and LEDs would be the next generation after that. Philips has a division called Luma Led, which makes high brightened LEDs and we’ve now put the Philips library of LEDs into PSoC Express software.

What that means is one of our customers who wants to use Luma Led can drag and drop an icon for a given Luma Led product, plug it into a PSoC design, and PSoC will comprehend what it’s driving and create all the software and drivers to drive it.

In addition to the efficiency I mentioned a minute ago, the other reason that this is important is that one of the things holding back LEDs in the market right now is that they are very non-uniform in manufacturing, though the way to solve that problem has traditionally been to create bins, so what happens is Luma Leds creates dozens of bins of products, meaning two LEDs that are supposed to be blue will be different brightnesses from each other and slightly different colors from each other.

If you want to match color and brightness perfectly, you have to supply a specific bin and you get charged a premium for that.

What happens with our software is that you can grab a bin, the machine knows what bin it is and it knows what correction factors are required for it. So this is an example where you can embed intelligence to take a complexity away from your customer in a new market which is growing rapidly, the LED market.

We shipped our 20 millionth wireless controller, 2.4 gigahertz, and we call it wireless USB. We also have another product called programmable radio on-chip, PROC, which is in fact a PSoC chip plus a radio. We now have 200 design wins on the radio and we are starting to get significant traction as we combine it with PSoC.

SunPower, as you saw in the report, had a great quarter. They announced their first plant is running at full capacity. They announced the opening of their second plant. It’s a massive building, 10 acres under one roof. It is three times -- twice the size of the first plant and it will be three times the output, doing 330 megawatts. The first plant is rated at 100 megawatts and I think it drifting up as they increase the efficiency of the plant. But this will make their total capability go up by a factor of four, relative to what it is now.

SunPower signed a continuous supply agreement with Hemlock Semiconductor. That’s a polysilicon and ingot company, one of the largest in the world. As you know there is a shortage of silicon in the world right now and SunPower is signing agreements to guarantee supply.

They also got a big win where they are installing solar systems in 28 Macy’s stores in California. They also got what I consider to be a premier win in Santa Rosa, California, Adjulant, which is showing its green in a very visible installation, a one million watt installation in Santa Rosa.

We sampled our first non-volatile static ram device. A comment on that; a non-volatile static ram is a static ram, but it is a static ram which can store data when the power goes off. That is a big deal. It combines the best features of static rams, which are fast and have excellent power speed product, infinite endurance with non-volatile memories, which can store data permanently even when the power is off.

We think over the next few years, this will grow slowly and become a very high profit margin generator for us. We are also planning to combine non-volatile static rams with PSoC to make systems that can sense signals and store information when the power goes off. So in effect, an electronic black box.

Finally, we sold our network search engine, or CAM business. This was a business that we had with Cisco in the second half of -- the CAM business we’ve been in for years, both to NetLogic, got $12 million for it, and sold inventory. We are now 100% out of the CAM business, which never made money for us and will help us become more efficient in producing gross margin and profit in the future by refocusing those resources.

Those are the highlights for the quarter, technical and business. We’re ready for questions.

Question-and-Answer Session

Operator

(Operator Instructions) John Barton, you may ask your question and please state your company name.

John Barton - Cowen & Co.

Thank you. T.J., with all the PSoC design wins under your belt, and I realize it’s touch to talk about averages, but can you give us a sense for the typical design in cycles as far as how long does it take, time to revenue, revenue per design, and again, if you want to vary from averages, talk about the extremes, that would be great too.

T. J. Rodgers

Let me first answer generically. Our foam is quantified mathematically. We have a model which has seven parameters in it. It starts with an opportunity or as we call it, an active if the customer engages with us in some way. There is then a delay from an active to a design win, and that delay is customized for every single product. We have about 20 different models, the products are very different.

We then have a percent probability of turning an active into a design win. We then have a second delay from a design win to revenue, and you have the second probability which says even though you’ve got the design win, the customer’s system didn’t work.

So you suffer from an active, you suffer two delays and two yields, if you want to call them that, until you get some money. And then money is -- I think it is six quarters for PSoC. Six quarters for a design win and you get so many units in a quarter. So we model each design win that way and we have models for each of our product lines.

For example, you might imagine rams are different from PSoC. So the details of that model are as follows. Chris.

Christopher A. Seams

T.J. gave away I think some of the yields. The latencies, John, from going from the active stage to the design in stage are typically between two and three quarters, more towards two, and going from the design in stage to the first revenue stage, which we would call a design win, are on the order of two to three quarters again, right in the middle.

T. J. Rodgers

So if you add those two together, you get five to six quarters. So the results we are reporting for PSoC today actually got created, that is, we knew about them, we knew about them for sure three quarters ago and we had a very good idea mathematically over a year ago.

A year ago when we were saying PSoC is going to be a big deal, the numbers weren’t all that great but this is why we knew it. And now in this quarter, we’re saying -- did we set a record in this quarter? Yes, we set another record for PSoC design wins and that makes us confident for 2008. That’s why Brad’s talking about it, unless there’s a problem in 2008, it will go up.

And it’s completely different and better than the life I used to live in the ram business, where you literally in the middle of a quarter could turn from everything is great to everything is awful, and surprise investors.

This long cycle time requires patience but it is also very stable.

Christopher A. Seams

One added comment, John; as we expand into the longer cycle markets, like automotive or industrial, and even some of the wireline communications with PSoC, those latencies or cycles push out somewhat, but the wins live longer as well.

John Barton - Cowen & Co.

Sure. And basic revenue for PSoC in the last quarter?

Norm Taffe

We don’t break it out, basic revenue. Hey, John, this is Norm. Frankly, in the last call, you could reverse calculate it and there’s the growth in this press release, but I can tell you that year over year, we expect to grow more than 50% in PSoC off of last year’s total number, and we did set, it said in the press release, a revenue record again this quarter.

T. J. Rodgers

Two-hundred million rate and growing at 30%-ish per year right now.

John Barton - Cowen & Co.

Thanks. And just real quickly, T.J., an update on the two fabs as far as utilization? Obviously you are not spending much on CapEx. What are your visions now with respect to how long those fabs remain and potential exit strategies, please?

T. J. Rodgers

Shahin Sharifzadeh, our VP of fabs, is in China right now, so I would ask Ahmad Chatila, who runs MID to answer your question on fabs. Ahmad.

Ahmad Chatila

Our utilization of the fab are excellent. They are in the 88% to 90% range. We continue to see strong utilization going forward. We are looking at options for Fab 2, the Texas one, because it is older generation technology and we’ll keep you posted on that one.

Our Minnesota fab, on the other hand, is leading edge technology and we will continue to utilize this for a long time to come.

John Barton - Cowen & Co.

Great. I’ll leave it at that. Thank you.

Operator

Thank you. The next question comes from Chris Danely.

Chris Danely - JP Morgan

Either T.J. or Chris, can you guys just give us your sense of how the end markets are doing out there? And are you worried about any sort of double ordering?

Christopher A. Seams

Let me answer in reverse order; double ordering I’m not worried about. I made a comment about lead times for us are in about the six to eight week range, which doesn’t precipitate those types of actions.

End markets for us, I’ll comment on. I’m probably not the best barometer of the overall end markets because we are penetrating some very big markets with some very new and exciting products, so we don’t get a true gauge on the overall market. We get a gauge on our penetration into that end market.

So with that said, my comments in the wireless handsets, that is a great market for us. We had a very good Q3 and we see that continuing into Q4, as we penetrate not only with PSoC but with our high-speed USB and our Antioch offerings into the cell phones as well.

Consumer, we expected it to be strong. Second half is always strong for us in consumer across the board and it didn’t disappoint and Q3 and Q4 continues to look solid off of Q3.

PC and PC peripherals, I lump together. We’ve always had a strong peripheral heritage and now we’re penetrating the notebook portion of PCs, which as many of you probably know is the fastest growing and most exciting portion of the PC market.

We are penetrating that with PSoC with multimedia Capsense controls. That was a growth segment for us as well in Q3 and we see a solid Q4 ahead of us.

Two of our more traditional markets, wireless base stations, which have been in the news of late and for probably about a year now, we continue to see a mixed market out there. It varies by account, whether they are recovering or growing or not, and so we are seeing what the basic market is showing but for us, across the board it’s not been a growth market for us but we don’t see it declining either.

Wireline networks has always been a strong market for us from our heritage point of view. That remains healthy going forward and we don’t see anything changing.

Chris Danely - JP Morgan

Great, and you said overall book-to-bill was above 1. Could you give us the book-to-bill by segment.

Christopher A. Seams

I can’t give it to you by segment. I can give it to you by product line.

Chris Danely - JP Morgan

Sure.

Christopher A. Seams

CCD is 1.05; MID, with the image sensors counted, is 0.97; and Data com is 0.95.

Chris Danely - JP Morgan

Last question; Brad, how do you expect options expense to trend going forward?

Brad W. Buss

I think it will be fairly consistent for us. I can’t quite speak for SunPower, but I don’t see any real major changes there. We’ve got an annual process coming up but we’ve also got stuff that declines off, so we’re managing it pretty tightly.

Chris Danely - JP Morgan

Great. Thanks, guys.

Operator

Adam Benjamin, you may ask your question and please state your company name.

Adam Benjamin - Jefferies & Company

I just want to clarify. You guys threw out a lot of stats on the PSoC business. I think, Norm, you said it would be growing year over year greater than 50%, and T.J., you said it’s at a $200 million run-rate and 30% growth. So based on that and given what you did in CCD, can you clarify a little bit? Is it the $200 million run-rate was from Q3? And would you expect to grow sequentially in Q4? It would seem to indicate that you wouldn’t, given the year-over-year growth.

T. J. Rodgers

My comment of a $200 million run-rate and 30% growth was forward-looking, talking about 2008. The 50% growth rate quoted was growth rate in the quarter relative to last year, so as we get bigger, our growth rate is still high but it is tapering off from a doubling every year. That is why two different numbers, two different timeframes.

Adam Benjamin - Jefferies & Company

Okay, and the last piece of that, would you be up sequentially, from Q3 to Q4?

T. J. Rodgers

The answer is yes. Not as dramatically as we were in Q3 on PSoC but we do expect to set yet another record in Q4.

Adam Benjamin - Jefferies & Company

Okay, and then I know you guys talked roughly about the margin being better than the CCD group margin. Is that -- was that the main contributing factor in terms of mix for the CCD margin improving sequentially?

Norm Taffe

Actually, two factors; one, PSoC growing, the other one is also our USB margins are improving as well as some of the high-speed, higher margin business grows in handset applications. Both of those contributed to improving gross margins in CCD and we expect that trend to continue.

Adam Benjamin - Jefferies & Company

And just to clarify regarding DCD, you saw a dip down in gross margin, which you are generally pretty stable there. I know the NSC business is coming out. Can you talk a little bit about any changes in mix there? I know that West Bridge comes online in terms of the ramp there, so can you just clarify what was driving the margin decline and whether we can expect that to continue, or if that’s a function of just a temporary mix?

Dinesh Ramanathan

It is actually just a function of temporary mix. The NSC business, as you mentioned, was very low gross margin and we basically sold that business. And then the rest of this business comes in at margins that are less than the old traditional communications business, so it is essentially a mix that is actually driving the gross margins in that direction.

T. J. Rodgers

We shipped our first significant amount of West Bridge this quarter and we are in early ramp phase and the gross margins are currently not what they will be in a couple of quarters, so that dinged gross margins in DCD a couple of points.

Brad W. Buss

You can expect their margins to still run in the low 60s like we talked about.

Adam Benjamin - Jefferies & Company

Dinesh, just to clarify, your NSC business had been running about $5 million or $6 million per quarter. You sold off $2 million per quarter, so we should expect it to still run kind of cash-cowing into $3 million? You seemed to indicate that that business goes away, but you still have a couple million a quarter there, correct?

Dinesh Ramanathan

No, there is no NSC business at all, so it goes to zero.

Adam Benjamin - Jefferies & Company

Okay, thanks for the clarification.

T. J. Rodgers

A year ago, we sold the bulk of our NSC business and kept one product that the acquirer did not want at that time. In the last quarter, we sold the rest of our NSC business and it now will be zero going forward.

Adam Benjamin - Jefferies & Company

All right. That’s all I have. Thanks.

Operator

Tim Luke, you may ask your question and please state your company name.

Tim Luke - Lehman Brothers

Congratulations on your execution. I was wondering, on the bookings number, it’s at 78%, which is a little lower than the 86% you had last quarter and how do you see that and how do you feel about that level in general?

T. J. Rodgers

If you look at the graph -- I’ll go to it, the beginning of quarter booked is the statistic you are talking about.

Tim Luke - Lehman Brothers

Yes.

T. J. Rodgers

The number when we’re having bad times is 60% and the number when we’re having good times is between 65% and 80%, and above 80%, the fact is we’re typically in trouble, scrambling and delinquent on orders. So 78% is a very good number, meaning we’ll have 22% of the quarter to turn in the quarter for us and it is just normal noise.

Tim Luke - Lehman Brothers

Just on the -- do you have the book-to-bill by business unit?

Christopher A. Seams

Let me just repeat it -- it is 1.05 for CCD; 0.97 for all of MID; and DCD is 0.95.

Tim Luke - Lehman Brothers

And if you had shared what the sequential improvement was in USB and whether you could just remind us of what percentage range that is of the overall CCD business. Chris, I don’t know if you had a comment on the overall linearity of how the business has proceeded through the calendar third quarter.

Norm Taffe

The percentage increase on the USB business, which I want to highlight actually includes our wireless as well, and we include that in the same business, which has grown strongly. So wireless USB and USB business grew very healthy in Q3, around the order of 20% quarter over quarter. And it represents roughly 40% of CCD.

Christopher A. Seams

Tim, your question on linearity, it was right in the slot from what we expected and what we’ve seen historically.

Tim Luke - Lehman Brothers

And then, any commentary on West Bridge in terms of how you’ve seen it develop and traction with large OEMs? I think you’ve talked in the past about how one Midwestern handset guy has been pretty -- has shown interest in the product and how is that developing in terms of broadening out the customer set?

Dinesh Ramanathan

So the Midwest customer is actually now putting phones in the market with our parts in it. We also have another North American customer that will put phones into the market, plus we are working with pretty much all the top tier handset customers that will either put phones in the first half or in the second half of next year. So we are seeing fairly broad adoption across the customer base. Actually, interest and adoption across the customer base that we have for that particular product.

Tim Luke - Lehman Brothers

So that business could be what sort of percentage range of DCD, perhaps in ’08 and ’09?

Dinesh Ramanathan

It could be anywhere between 20% and 30%.

Tim Luke - Lehman Brothers

T.J., [inaudible] lastly, commentary on the general environment and following up on Chris’ question in terms of how you perceive the outlook, maybe visibility the beginning of next year?

T. J. Rodgers

That’s for me?

Tim Luke - Lehman Brothers

Yes.

T. J. Rodgers

The environment is not really hot right now. We’ve been -- 2007 was not a great year in general in the market. It was okay. It was not a disaster. We were very happy that our divisions all made it through every quarter making money all year long, which is very different than old Cypress was when we had a tough year.

Right now, the market is the same as 2007. We are doing well. We’ve got a lot of new products. That’s driving our momentum but I am looking forward to better markets into the future. This is not a great semiconductor environment. It’s stable. It’s not bad. It’s stable at a B-minus kind of level.

Tim Luke - Lehman Brothers

Thanks, guys.

Operator

Sandy Harrison, you may ask your question and please state your company name.

Sandy Harrison - Signal Hill

As far as the PROC, you had talked a little bit about it starting to come into its own, and given the timeframe you had talked about for your PSoC adoption of say three to six quarters from sort of design win to revenues, the PROC’s been out now going on three to four quarters, as I recall correctly. What sort of layering do we see PROC coming in? And if you could spend a little time reminding us of the ASPs here and what sort of applications those could play in?

Norm Taffe

You’re right. We came out with -- we introduced PROC about three quarters ago. It hasn’t contributed significantly to our revenues yet, although I think we did break our first $1 million quarter, so it’s off the mark. It is getting designed in quite heavily. It is primarily just what you’d expect -- places where people want to combine a radio and a micro. We actually have quite a bit of applications in toys, toy applications. We have quite a bit in the industrial space where people are trying to do wireless sensing kind of applications.

It is very much a many-broad customer type market, like PSoC will be over time. We’ve broadened the customer base, as we’ve broadened the customer base for PSoC, so it is very broad in that perspective.

Therefore, even though it incorporates our wireless USB radio, it does have better margin portfolio in general than USB products, more consistent with the margins we have in PSoC.

Sandy Harrison - Signal Hill

Okay, and again, where do you think that -- are we seeing another, consistent with your sort of path to revenues, another couple of quarters before it really starts to drop in meaningful levels? Are there some particular design wins out there that give you that confidence or is it just sort of a rising of the tide inconsistent with your PSoC?

Norm Taffe

I think it’s mostly rising of the tide. There’s a couple of big wins that could make it even more special in terms of percentage, but one of the ways we look at PSoC is a core device that allows us to have a variety of interfaces. We have PSoC with a USB interface, we have PSoC with wireless interface. In general, the broad strength of PSoC, this is just another way in which we exploit that in a select set of applications that use wireless technology.

Sandy Harrison - Signal Hill

Okay, and just a commentary, kind of combining T.J.’s comments on the industry and where it stands on an earlier question, as well as some of the inventory and seasonality that the industry is expecting, given the lead times being relatively short but hearing out there that fab capacity is tight, not only now with you guys but as well some of the merchant providers, do you think that there is any risk potentially that on a strong Q4, that inventory and availability of products might become an issue out there in the broader markets?

T. J. Rodgers

Let me comment on that. One of the reasons we are starting to take off and have our fall through gross margin go right down to the bottom line and our EPS accelerates, is that we’ve got Fab 4 full. And in the future, instead of adding equipment to Fab 4 and doing the old game or doing Fab 5, we’ll be buying wafers at foundries outside and keep Fab 4 full.

So back on your comment, is capacity full? Our internal capacity is pretty much full and we are now overlapping in the foundries and our foundries are reasonably full. And we have contractual arrangements to get the wafers we need, but it is true, despite what I call the B-minus demand environment, it is also true that worldwide, fabs are pretty full right now and any surge would lead to a shortage of product.

Sandy Harrison - Signal Hill

And then lastly, sort of taking a step back and looking in the crystal ball, T.J. and Brad, you guys have done a heck of a job financially changing the company, both from a fundamentals perspective, from a manufacturing as well as managing your costs by outsourcing and others.

What’s next? You guys, again you’ve done a great job and in the spirit of Wall Street, what have you done for us lately kind of question? What could we look for and what could be some potential next steps that you guys are looking at internally?

T. J. Rodgers

We’ve still got some more clean up to do inside. There are a couple more small-ish business units that aren’t all that visible to you that we are going to sell the businesses and basically transfer the people on to our main event.

We have some cleaning up to do in manufacturing and that will show up in terms of better gross margins for [inaudible], price, we’re certainly revamping all of our manufacturing methods in the company to reduce our costs, take advantage of the very low cost Chinese foundries we’ve transferred our technologies into.

With regard to initiatives, things that we can talk about, we have a new internal start-up called Cypress Systems. They recorded first revenue this quarter I think in the order of $100,000. I don’t remember the exact number.

For a long time, I’ve looked at making chips, for example, radio chips and selling the radio chip for -- PROC, let’s say, for $2 or $3, and it allows you to hook up something, let’s say a temperature or humidity sensor to a radio, have the radio transmit to your computer, go into another [download] that’s got a couple of Cypress chips in it and take remote data from a radio from any sensor you want into your computer.

It has always aggravated me that we get a couple of bucks on each end for that kind of amazing capability and yet when I buy stuff like that personally, weather station type stuff, I pay anywhere from $50 to thousands of dollars, depending upon the number of channels that come in.

So we started Cypress Systems to make small systems. We obviously face the potential risk of some competition with our customers. We’re avoiding that but basically, we’re taking our products and making systems which sell for hundreds of dollars instead of single-digit dollars, our ASP being $1.29 in the company.

We’ve been working on the company for two years, a year-and-a-half. We invest about $2 million a quarter, so it takes a penny per quarter out of our earnings and that will become a million dollar per quarter business I think late in 2008. We have a chance to hit some homeruns in that business.

We’ve been working on that for a couple of years and it just now will become visible to investors. I’ll let Terry Sim, the President of Cypress Systems, come in either next meeting or meeting after that and describe what he’s doing.

Brad W. Buss

If I could just add a couple of things, I appreciate your comments. I mean, the team has done a lot but I think the big thing is the financial impact, the market expansion we’re going into, we’re still leaving first base. We are going into these multi-billion games we’ve never been in.

So even if semis don’t grow, we’ve got a ton of CAM expansion across the board that is just starting, so that’s driving revenue. The gross margins we’ve talked about between the switch to programmable product, more impacted in our COGS structure. We only see that moving up into the right as well, so we’ve got -- and then plus the new products we are currently developing, next gen of PSoC and a handful of other things, we’re cranking up a pretty heavy organic revenue stream. And to T.J.’s point, we’re dropping more and more of the incremental dollars to the bottom because our cost structure has been streamlined.

We’ve got years of that ahead of us, so maybe you guys should start looking at the stock value appropriately.

Sandy Harrison - Signal Hill

So we’ll look for a bricks wireless sensor in the winemaker magazines coming forward?

T. J. Rodgers

There is actually a company that is outside of Cypress' family called Provena. You ought to check them on the web. They make a -- I licensed them the technology I developed in my own winery using PSoC technology. You can buy a fermenter, a real fermenter that does real, professional fermentations to put in your house to put up the web and actually be connected to instructions on how to make wine properly and you can look at the parameters in the fermenter. It is an RF connected Internet appliance, and if you want to get an example of somewhat of the more bizarre things that you can do with PSoC and radios, go take a look at Provena on the web and given that you guys are all pretty much yuppies in your jobs, you guys maybe all want to buy one for your house and start making your own Cabernet.

Sandy Harrison - Signal Hill

All right. Thanks, guys.

Operator

John Pitzer, you may ask your question and please state your company name.

John Pitzer - Credit Suisse

Thanks, Credit Suisse and this is Ahmad. Just taking a high level look at PSoC, when you see growth next year, do you see most of the growth coming from an expansion of the market or do you see yourself taking market share from some of the traditional MCU guys?

Norm Taffe

I think we’re entering a very, very large TAM which grows at a growth rate of 7% or 8% a quarter and we’re going to grow a lot more than that, so clearly we -- starting [inaudible], excuse me. So clearly we will continue to take market share in the space.

I do want to highlight that our TAM, and we take pains to point this out, really extends well beyond just the microcontrollers. We are -- we really find ourselves often replacing four markets -- a microcontroller, essentially ASICs, in the lower end of the ASICs base; programmable logic, because there is a significant amount of programmable capability in the product; and then importantly, analog. And in all four of those markets, it’s an enormous TAM which we’ve just barely touched, and we see our value being ability of a flexible integration of all those technologies.

So in the end, do I think all four of those markets will grow anywhere near where -- no way, but I expect us to grow very, very quickly by taking share within that space and there is plenty of TAM for us to grow into.

John Pitzer - Credit Suisse

And then, on the non-volatile SRAM, do you see that as a market in and of itself, or is that really the way we should think about it, as some type of an attach market with other products? What is the market potential there that you see in 2008?

Ahmad Chatila

You know, NVSRAM goes into what is called battery-backed market, battery-backed ram market, as well as [inaudible], NRAM, and NVSRAM, and that’s a multi-hundred million dollar market, as well as the low-powered SRAM plus a battery that is homegrown solution. So if you go to some companies, they buy the battery themselves and they buy a micro-powered SRAM from us.

So actually, if you look at the overall potential, it’s a $1 billion market. With our solution, it is the only one that you do not have to us esoteric material that is difficult to manage and attack that market.

So initially, we are just trying to develop our market there in the $1 million to $2 million a quarter, and we will get there in short order. But our aspiration is to grow and go after the $1 billion market. But then as T.J. said, we are going to add PSoC to that and that will grow our solution even further and make it a proprietary product.

T. J. Rodgers

Let me give a little more color on the last two answers. The answer is yes, we are taking market share in the microcontroller business. If you’ll remember the pitches I was giving in 2000 and 2001 when we launched PSoC, we were planning on getting the microcontroller business. We looked at one of the more attractive competitors in that market, which is Microchip, and the problem we had in trying to enter the market is we had 10,000 products.

So PSoC actually got invented to be a programmable microcontroller where we could take over the function of approximately 5,000 Microchip products with only six chips, where the microcontroller would be in the center but the peripherals, the digital and analog peripherals, would be programmable. So you could spin your own microcontroller and we actually in the beginning mapped our offering on to Microchip’s product line with the concept of dislodging them in the market.

It turned out that wasn’t the best strategy but it put us in the right place. By emulating their capability in the marketplace of 5,000 parts, we basically created a capability that stood on to itself. With the system on chip, you could do stuff you couldn’t do with a fixed function microcontroller.

Therefore, one way to look at our market is that we are taking share because we can give you a microcontroller and we can integrate analog functions that you would have to buy otherwise from Linear Technology and digital functions you might have to buy otherwise from [Alliance Semiconductor], and our big sell is not we are going to take that microcontroller socket. Our big sell is put a circle on your board, look at the analog and digital functions surrounding your microcontroller, they’re on board our chip. Our chip is therefore more cost efficient and more flexible.

By the way, you can change the actual hardware in a microcontroller five times in the last month before production and not get shafted because you are actually changing hardware.

So the IT base of our design maps on the microchip but we’ve really created a new category of product -- a programmable system on chip and it now has a life of it own and we are not really competing head-on in the microcontroller business. We are taking greenfield system design.

The second comment on non-volatile static ram, I told you what it was before. The static ram, which is the most convenient, efficient and power friendly of all memories, with the ability to store data when the power goes down, and that’s to store data without using a battery. The energy required to move the data from the static ram cell into the non-volatile cell and their mapped 4 million of each, so there is a non-volatile cell directly next to a static ram cell inside the memory.

The energy to do that is stored in the capacitor, so it doesn’t require a battery. So any system where there is a battery-backed monitoring system, so when the system goes down, you can remember what happened and that’s basically any router, any blade server, any system with which needs to recover from a crash currently uses battery back-up ram. The problem with batteries is that they are very unreliable, relative to solid-state semiconductors. They are difficult to solder on the boards because the battery may or may not be able to take the temperature, and that’s gotten worse when we go lead free, which is higher temperature.

So the non-volatile static ram I think is a major initiative and I think when we combine it with PSoC, it is going to create a whole new class of product. This is a product which can sense the environment.

I’ll just give you one example. The guy at EMC, for example, and I’m creating a hypothetical example, let’s say the guy at IBM walks up and say my blade server crashed. Mr. PSoC, what happened? Mr. Non-volatile PSoC? And the non-volatile PSoC says I was monitoring four power supplies every 200 milliseconds before the crash and what happened was power supply two started to fade, and furthermore, the temperature on temperature sensor one started to go up. This is what was happening in your system before it crashed. And that data can be measured and stored in PSoC. You can take it back out after the event and go back and do root cause corrective action on what’s wrong with your system.

That function, which I call black box, I think is going to become more and more pervasive as people demand in effect perfect performance from systems. And therefore, the combination of that kind of very special memory with PSoC in the future, I think it will be a slowly rising and as somebody said earlier, the tide rising kind of business, but a very important, high quality business that we will build over the years.

John Pitzer - Credit Suisse

If I can just sneak in one last quick one, and that’s about valuation. Can you just update us on your philosophy for monetizing SunPower? I mean, do you see that as a 2009 timeframe or I sit -- what are the factors you think when looking at that stake?

T. J. Rodgers

First of all, I would like to thank you for asking that question without threatening my job, as in prior meetings. I expected the question would get asked, so I got a phone call to their lawyers and accountants last night to refresh myself on the dynamics and I also read the transcript from our January 25th meeting to make sure that it was accurate, and I had that in mind when I answered the question.

Bottom line, that transcript is on the web and when I read it, it was a very clear answer and it was exactly where we are right now. It sounds like a blow-off when I say we continue to evaluate options. That doesn’t mean like the government continuing to study and do nothing. It means we continue to evaluate our options and right now, there are no options which will make shareholders better off. And as a 3% shareholder myself, I am very interested in making shareholders better off but right now, we see no deal that is better than status quo.

And the details, there’s four options for separation and the details of those four options, which take a while to go through, haven’t changed since the January 25th phone call, which is on the web.

John Pitzer - Credit Suisse

Sounds great. Thank you.

Operator

Thank you. Our next question comes from Srini Pajjuri. You may ask your question and please state your company name.

Srini Pajjuri - Merrill Lynch

Just quickly on the PSoC, obviously you had a lot of success and it’s been a very high visible product for you. I’m just wondering if you are seeing anything else out there that is competitively coming close to this product?

Norm Taffe

As far as have we seen anything else, T.J. I think said it well earlier, where we don’t see anything that’s really a direct competitor to the product. I will say that I am confident that our success is being noticed and we do see people integrating analog features on maybe a greater scale, and specifically -- and we’re constantly cautious to make sure we don’t, we are looking closely.

Haven’t seen anything that we see as directly in response to our product or like it and we are certainly working on the next generation very actively to extend our position there. But it wouldn’t surprise me if something like that arrived. I just haven’t seen anything yet.

Srini Pajjuri - Merrill Lynch

Okay, great and then one question on the SRAM business. That business has been fairly stable and it’s been very profitable, it looks like. I’m just wondering, what’s going on in that business that has changed there? Has it become less cyclical or if you could talk about maybe what’s your outlook for the next couple of years, how you expect that business to trend?

Dinesh Ramanathan

We expect in the next couple of years to continue to gain share and continue to improve our gross margin.

Yes, the business has changed a lot -- less competition and I would say a lot more difficulties penetrating customers. It is not easy just to design an SRAM and throw it at customers. And we have a very broad portfolio and everybody likes to be with us because we can give them from a 16K SRAM to 72-meg QDR. So I expect our gross margins to continue to improve step by step by step.

And our revenue outlook, I can’t tell you a lot because, dependent on the market growth. There’s not much there so between our market share growth and market decline or flatness, we might see a little bit of an up, but not much. But the profit will increase over time.

Srini Pajjuri - Merrill Lynch

Okay, and then if -- let’s say if wireless infrastructure comes back at some point, does it really help your SRAM business or is it less of an issue?

Dinesh Ramanathan

Let me answer that -- when the market on wireless base stations crashed, it didn’t crash at every account at the same time. So actually, you start to see it in some accounts and then later on in other accounts in other quarters. And in the same fashion, some accounts ramped before others, so I do not see a big impact going forward on that.

Srini Pajjuri - Merrill Lynch

Okay, and then more of a longer term question for T.J.; T.J., as you look at the SRAM business, obviously it is doing well right now but to me, as PSoC and other proprietary products become bigger and bigger, I’m wondering if there is a real synergy between your SRAM and your other businesses. Longer term, does it make sense for you to stay in this business?

T. J. Rodgers

Very short answer; in the next three years, SRAM is wired into Cypress. At a point seven years from now, might Cypress not have an SRAM component? That’s a decision we’ll have to make longer term.

In the short term, SRAM does two important things for us. I look at SRAM as being a conveyor belt that takes depreciation money out of my fab and put it back into the bank, so with SRAM, we always have our fab full. We get the depreciation and cash flow out of our fabs. We’re still depreciating -- what’s our depreciation rate?

Brad W. Buss

$19 million, $20 million a quarter.

T. J. Rodgers

$19 million, $20 million a quarter, we want to do that. The second big thing SRAM does for us is that if you look inside a PSoC, you will see a very dense SRAM cell. Not a typical, half-assed logic SRAM cell, which many of our competitors have, but a real honest-to-god full-fledged memory cell, because I’ve got real memory designers in the company.

So as long as SRAM is an inherent part of our product, it is also nice just to be able to ask the SRAM guys to give me a macro for my other product and I know I will get it, I know it will be robust, it will be reliable, it will be radiation tolerant. So for those two reasons -- and let me give a third reason, too.

The third reason is that the SRAM business is making a nice profit right now and we believe it is going to get more profitable and more stable over time because the history has been so bad it squeezed out a lot of competitors. And therefore, it’s a great cash flow and profit generator that we can use to, for example, invest in Cypress Systems, which is costing a penny a quarter, and still put enough money on the bottom line that you guys don’t gripe that our R&D is too high.

So for right now, the longer the core of the company, Cypress used to be an SRAM company. We’re not now. We’re a programmable products company but it certainly does add a lot of value to Cypress and we don’t anticipate any changes.

Srini Pajjuri - Merrill Lynch

Thank you.

Operator

Thank you. Glen Yeung, you may ask your question and please state your company name.

Glen Yeung - Citigroup

T.J., you characterized this year, ’07, as kind of a B-minus year. When you look out into next year, again thinking industry, do you think we are going B-plus or are we going C?

T. J. Rodgers

I’m a bad person to ask. If you look at my history, I am usually buying steppers directly before a crash. Chris.

Christopher A. Seams

I think that current industry pundits and leading services are forecasting or predicting about 3%, 3.5% growth for ’07. I think they’ll get the right answer as they get the numbers out of December.

When they started the year, they predicted about 9.5% growth for the industry. The current roll-up for next year says about the same number, about 9%, 9.5%. So according to the industry forecasters, it says it’s going to be up versus ’07 but as T.J. said, they are probably not any better than he is.

Glen Yeung - Citigroup

Okay, and there’s nothing that you can detect in your own business that can give you any visibility, partially because you are gaining so much share in your products, I guess, it’s a little masked --

T. J. Rodgers

We’ve said it before -- no matter what the end market does, we’re going to take share and plan on growing.

Glen Yeung - Citigroup

Brad, just thinking about, just researching the semi business, can you talk about where you think margins are going to, gross margins, that is, really over next year is what I’m thinking about? And can you remind what gross margins in the semi business were in the June quarter?

Brad W. Buss

Sure. The June quarter, you said?

Glen Yeung - Citigroup

Yes, the one -- last quarter.

Brad W. Buss

Yes, hang on. I’ve got the spread there in the press release. It was total semi gross margin, 44.9, and then they went up to 47.2.

Glen Yeung - Citigroup

And how do you think we look next year?

Brad W. Buss

They are going to go up. I expect them to go up actually almost every year. There’s a lot of factors to it, right, and we’ve talked about the switch in the business. T.J. went into all of our programmable stuff is well north of 50%, number one.

The value selling that we’ve been doing as a company, again a big switch from memory roots to programmable roots, is just going into all the new designs that won’t even hit revenue yet. That provides enough lift.

The more business that goes through foundry, which is where our track is going, provides more benefits to gross margins.

I think if you add all of them together, it bodes very well for gross margin.

We are in the midst of looking at our 50-30-20 model. I am pretty comfortable we’re going to hit that next year and we will be upping the 50% and I am hoping to up the 20% target for our next few years of future.

Glen Yeung - Citigroup

Do you think upping the 20 is purely a function of upping the 50, or do you think you can lower the 30?

Brad W. Buss

I think it will be mostly on the 50. I mean, there will be some room potentially, but we are trading off op-ex dollars. I mean, we’ve done a very good job of decreasing op-ex and reallocating, even within the dollars we have, to the high growth areas and I think you’ll see us continue to do that.

We’re going to need to continue to invest in FAE and apps engineers to keep seeding these big markets that we are going after. But I think overall, you are not going to see the cost structure get too crazy.

And to T.J.’s point, SLM and Cypress Systems are actually a drag on op-ex right now and aren’t contributing to the bottom line. They both will continue and add revenue and dollars to offset costs going forward as well.

So I am pleased with where the cost structure is and the leverage will come, as we mentioned, from increasing revenue and dropping a high majority of the incremental GP dollars to the bottom line going forward.

Glen Yeung - Citigroup

Do you have any feel for what you think capital spending will be next year?

Brad W. Buss

I think 50 is probably a good number you can model in for the future. It will be plus or minus, depending on timing but I think that’s a good number.

I mean, even if you look at a fab-less company, as a percent of revenue, that’s a pretty low percent of revenue, so I am real pleased with where we are going on CapEx and I think you start adding that, plus the incremental earnings, you see tremendous free cash flow being generated going forward.

Glen Yeung - Citigroup

That’s really where I was going with this. We’re seeing accelerating revenues, expanding margins, falling spending -- it looks like we are going to have pretty interesting cash flow next year, so any thoughts about uses of cash, and sort of traditional uses -- buy-back, dividends, et cetera, but also thinking about -- you’ve been divesting businesses. Any chance you may be looking to acquire anything?

Brad W. Buss

I think from a capital structure standpoint, I’ve been pretty vocal that I’m not a big believer in a ton of cash piling up, so we’ll put it back to work in some of the traditional things. We could explore dividend longer term for sure. I think buy-backs are a real big part of our future and we’ll -- we definitely always continue to look at acquisitions if they make sense.

But like I said, we’ve got a tremendous amount of organic revenue growth. If there is something that could complement our strategy, fill out a product line, or some kind of key core, but I think -- you don’t know. There is a lot of stuff in play, though. There’s a lot of guys that I think are trying to figure out scale and where they could play and if something presented itself at the right synergistic look and valuation, we would look at it.

But I wouldn’t say personally it’s a core part of what we need to do because we’ve got a lot of good things going on.

T. J. Rodgers

We had an acquisition binge in the 2000 timeframe. Some of the stuff worked. For example, the radios we talked about were a combination of two small companies we acquired. Some of the stuff didn’t work; for example, the CAMs that didn’t work out and we’ve now gotten rid of them.

So acquire to get a new capability is probably not in the cards. We’ve got enough interesting stuff to work on in a $10 billion microcontroller market to grow what we need to grow.

If we considered an acquisition, which to me is something I do if it’s compelling only, to consider an acquisition now, it would have to fortify a position in the market we’re in and make sense, not off in some new thing.

Brad W. Buss

And just to tag on about the internal start-up concept that T.J. has pioneered here has paid very good dividends. You’ve seen it in PSoC. We’re seeing it Systems. We’ve had a quasi-impact of that with SunPower and SLM starting to come into fruition, and I think that’s a good strategy that we could keep doing going forward, but to build into the areas we want on a lost more cost-efficient and focused basis.

Glen Yeung - Citigroup

Yes, that’s a good approach. All right, thanks a lot.

Operator

Thank you. Douglas Freedman, you may ask your question and please state your company name.

Douglas Freedman - Amtech Research

Thanks for taking my question. A lot of them have been asked. Brad, clearly cash flow is improving, profitability improving, any help on tax rate going forward? I know there are some NOLs left there. Any update you can offer?

Brad W. Buss

Well, I gave you the guidance on the tax rate. Pretty low -- I mean, we’re going to run at 8% to 9% in the semi business, I think far greater than the vast majority of companies so I don’t see the rate changing much.

I mean, we do have a substantial amount of NOL, that the only way they are going to get eaten up would be a SunPower sale of stock and we have a ton of tax coverage going forward.

Douglas Freedman - Amtech Research

So continue to hold the overall trajectory that you -- I believe you talked about a 15% rate for next year. Does that sound about right, consolidated?

Brad W. Buss

No, I think it will be lower. I mean, I haven’t looked totally into that but I think SunPower’s rates came down. Probably consolidated, 11%, 12% but again, I’d caution you to really do the rate based on each bucket, depending on how you are forecasting revenue and income and then let it blend naturally.

Douglas Freedman - Amtech Research

All right. You just made some --

Brad W. Buss

-- look at the two pieces separate in your model.

Douglas Freedman - Amtech Research

Perfect. I guess I’ll leave it there. Thanks, guys, for all the detail. I’ve gotten a lot of the questions answered.

Brad W. Buss

Thanks, Doug and thanks for your very intuitive research report.

Operator

Our last question comes from Thomas [Domek]. You may ask your question and please state your company name.

Thomas Domek - Goldman Sachs

Thanks very much. My question’s already been answered and congratulations on the quarter.

Operator

Thank you, sir. At this time, we are showing no further questions. I will turn the call back over to your for any closing comments.

T. J. Rodgers

Thank you very much. We just reported the third quarter, $450 million, a record, $0.26 above the Street. We’re happy. Thank you for calling in.

Operator

Thank you. This does conclude today’s Cypress Semiconductor conference call. Have a nice day.

Copyright policy: All transcripts on this site are copyright Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.