Good day, everyone, and welcome to Microchip Technology fourth quarter and fiscal year 2006 financial results conference call. As a reminder, today's call is being recorded. At this time, I would like to turn the conference over to Mr. Gordon Parnell, Microchip's Chief Financial Officer. Please go ahead sir.

Thanks, Justin. Good afternoon, and welcome to everybody. During the course of this conference call, we will be making projections and other forward-looking statements regarding future events of the future financial performance of the Company. We wish to caution you that such statements are predictions and actual events or results may differ materially. We would refer you to our press release of today as well as our 10K for the fiscal year ended March 31st, 2005 and our 8-K current reports that we have filed with the FCC that identify important risk factors that may impact Microchip's business and results of operation. In attendance with me today is Steve Sanghi,, Microchip's President and CEO. I will comment on our fourth quarter and fiscal year 2006 performance, reviewing geographic data and discussing balance sheet and cash information. Steve will then give his comments on the results, outline our guidance for the June quarter, and update other matters regarding our business. We'll then be available to respond to specific investor and analyst questions. Let's get going. Net sales for the March quarter were at record levels of $247.2 million, which were up approximately 5.2% from net sales of $234.9 million in the immediately preceding quarter and up approximately 18.8% from net sales of $208.1 million in the prior year's fourth quarter. Net sales for fiscal year 2006 were a record $927.9 million, up approximately 9.6% from fiscal 2005. On an earnings basis, our GAAP net income for the March quarter was at record levels of $75.6 million or $0.35 s per diluted share, an increase of 6.8% from non-GAAP net income of $70.7 million or $0.33 per diluted share from the immediately preceding quarter, and an increase of 34% from GAAP net income of $56.4 million or $0.27 per share from the prior year's fourth quarter. Non-GAAP results for the immediately preceding quarter exclude the effects of the one-time tax charge associated with the repatriation of earnings under the American Jobs Creation Act. On a non-GAAP basis, net income for fiscal 2006 was a record at $273 million or $1.27 per diluted share, an increase of 20.4% from net income of $226.8 million or $1.07 per diluted share in the prior fiscal year. On a GAAP basis, net income for fiscal 2006 was also at record levels of $242.4 million, or $1.13 for diluted share. That was up 13.4% from net income of $213.8 million or $1.01 per diluted share in the prior fiscal year. This marks our 62nd consecutive quarter of profitability, achieving approximately $1.3 billion in cumulative earnings for Microchip and our shareholders. Geographically, sales in Europe grew by approximately 18%; Americas grew about 9%;, and Asia fell by approximately 4%. Asia reflecting the seasonal factors related to the lunar New Year holidays. Asia continues to be our largest geography, representing approximately 42% of sales. Americas was 29.25% of sales, and Europe the balance at 28.75% of sales. This measurement for the three geographies is based on whether a product is delivered for manufacturing purposes per a customer's request. It doesn't represent where the design activity is taking place. Obviously, many designs in Americas and Europe migrate to Asia to access more cost-effective manufacturing opportunities. In the quarter, we established record gross margins of 60.1%, reflecting the propriety of product positioning relative to pricing, world-class manufacturing yields and efficiencies, and richer product mix. We are very pleased with the continued improvement in gross margins that we have delivered to the market. Over the last fiscal year, our gross margins have improved approximately 275 basis points, significantly improving our operating leverage. Our current gross margins have crossed an important milestone at 60% and are an excellent indicator that we can meet our long-term gross margin target of 62%. Operating expenses were 23.9% of sales overall in the March quarter. This compared to operating expenses of 23.7% in the previous quarter. Research and development costs were $24.5 million representing 9.9% of sales, and SG&A expenses were $34.6 million representing 14% of sales. The tax rate for the March quarter was in line with our guidance of 24%. Our tax rate for fiscal 2006, after adjusting for the one-time tax charge, associated with the repatriation of foreign earnings under the American Jobs Creation Act I mentioned earlier, was also 24%. Inclusive of this charge, the tax rate was 32.5%. The dividend we declared today of $0.215 was an increased dividend of approximately 13.2% sequentially, and an increase of 126% over the same quarter in fiscal 2005. During fiscal 2006, our dividend payments totaled approximately $120 million and the dividend just declared would represent an annual rate of dividend payment of approximately $183 million. Overall in fiscal 2006, we established new record levels in net sales, net income, earnings per share, gross margins and operating margins as well as obviously dividend payments. Let's turn our attention to the balance sheet now and go through some of that information. First of all in Microchip's inventory, our total inventory position at March 31st was approximately $115 million with days of inventory representing 106 days. This is lower by 5 days from inventory levels at the end of the previous quarter. In distribution, our inventory continues to be at very moderate levels in all geographies. As of the end of the March, our distributors held about two months of inventory. This is essentially flat with the prior quarter. Given the industry conditions and the level of inventories at distribution, we are comfortable with the overall level of inventories we have supporting our business today. Looking at our receivables at March 31st, Microchip's receivable days were essentially unchanged from the prior quarter. Overall trade receivable balance increased approximately $27.4 million from the end of December. The delta between the revenue growth and the receivable growth can be contributed to several factors. These include the fact that the strongest growth in the quarter came from the European region where customers' payment terms are generally longer than our other geographies. We saw an increase in dollar value of inventories at distributors, which is not recognized in net sales, due to our revenue recognition being based on sell-through criteria. Lastly, we had more back end waiting for revenues in Asia, which related to the lunar New Year in that particular geography. We certainly don't have any concerns regarding the growth of the receivables, more an indication of improving cash flows in the current period. In our overall cash position as of March 31st, Microchip's cash and total investment position was approximately $1.3 billion with $269 million of short-term debt on the balance sheet. During the quarter, we generated net cash flow from the business of $110 million, prior to the dividend payment of $40.5 million. With the adoption of FAS 115, our investments are reported differently than before. In our balance sheet presentation, investments are split between current and long-term assets based on the criteria of FAS 115. Cash balances will continue to be reported in current assets, and all three categories are included in our total cash and investment position of $1.3 billion. The borrowings on our balance sheet relate to transactions associated with the repatriation of foreign earnings under the American Jobs Creation Act. To complete the repatriation of $500 million, we initiated certain borrowings which were collateralized against investments that were held in the foreign locations, allowing investments to reach their normal maturity date. We expect these borrowings to be paid down in the next four quarters, as the investments mature and new cash is also generated in these particular locations. There is no net income effect during this timeframe, as the interest income on the funds repatriated essentially offset the interest expense on the borrowings. Looking at capital depreciation, our capital expense, our spending was approximately $33 million for the March quarter. Depreciation expense for the same period for March was $26.9 million versus $28.7 million for the same quarter last fiscal year and $26.8 million in the December quarter. Capital spending for the fiscal year was $75 million versus $63 million in fiscal '05. Depreciation for 2006 was $109 million versus $119 million in the prior year. We anticipate capital expenditures of approximately $80 million, and depreciation of approximately $115 million for fiscal 2007. Operating cash flows for fiscal 2006 continue to be extremely strong, reaching record levels of $450 million in net cash generated prior to the dividends of $120 million that I mentioned earlier and stock buyback activity of approximately $3 million. During fiscal 2007 we anticipate that net cash flow prior to dividends will be approximately $500 million. Microchip will adopt FAS 123R for fiscal years beginning April 1st, 2006 and based on investor feedback, Microchip's board has determined to use restricted share units, RSUs as the equity vehicle to incentivize our employees. RSUs will replace stock options as the primary long-term compensation vehicle at Microchip. Based on our analysis of FAS 123R implementation, we anticipate an earnings impact of between 7% and 8% during fiscal 2007. We will report earnings on a GAAP basis, as well as excluding the effects of the stock compensation costs, to allow investors and analysts to understand the adoption of FAS 123R. With that, I'll ask Steve to discuss the performance of our business, our guidance in the June quarter and update other business matters.

Thank you, Gordon. Good afternoon, everyone. Today I would like to reflect on the results of the March quarter as well as the fiscal year 2006, and then comment on the product lines, update you on our distribution strategy and on our fab manufacturing capacity, and finally discuss the guidance for the June 2006 quarter. So let's begin. I'm extremely proud of our continuing excellent execution in delivering yet another record quarter and another record fiscal year in every respect. We exceeded our original guidance in net sales, in gross margin percentage, and in EPS. Net sales were $247.2 million, which was an all-time record. Gross margin of 60.1% was also a record, up 40 basis points sequentially. Our operating profit was another record at 36.2%, and EPS of $0.35, exceeding our original guidance by $0.01, and it also set another all-time record. From a fiscal year standpoint, we achieved a record net sales of approximately $928 million in fiscal 2006. Gross margin was also a record at 59.4% and non-GAAP EPS was a record $1.27. Non-GAAP EPS excludes the effect of foreign earnings repatriation that Gordon talked about. Now I shall talk about the product lines. First the microcontrollers. The microcontrollers business grew 4.9% sequentially and was up 17.8% over a year ago quarter. Our flash microcontrollers business was very strong and grew by a robust 9.6% sequentially and up 42% over a year ago quarter. Flash microcontrollers now represent over 58% of our microcontrollers business. We also shipped a record 19,991 development tools last quarter and also a record 73,705 development tools in the fiscal year 2006. Development tools shipped in fiscal 2006 were up 38.3% from those shipped in fiscal 2005. We are continuing to experience strong design activity with our 8-bit flash microcontrollers, and we're seeing a significant number of new applications emerging which are utilizing our products. We are very well-positioned to continue to gain market share in this segment. Gartner DataQuest released a 2005 Microcontrollers Market Share Report just last week, which confirmed again that Microchip is continuing to gain market share in 8-bit microcontrollers. Of the top four suppliers who have a combined 52% of the 8-bit microcontrollers market, Microchip was the only company that grew year-over-year. The other three companies -- Freescale, Renaissance and NEC -- all experienced a year-over-year decline in 8-bit microcontrollers sales. Now, 16-bit microcontrollers. Our 16-bit microcontrollers business was up a whopping 46% sequentially in the March quarter, albeit from a small base. So far the growth off our 16-bit microcontrollers business is higher than that of our high-end 8 bit microcontrollers architecture, Pic18. The number of customers on 16-bit grew to 324 in the March quarter from 236 in the December quarter. Based on that metric, the customer penetration is proceeding as expected. We shipped 1,284 new development tools last quarter, on 16-bit for a cumulative total of 12,671 development tools supporting our 16-bit products. In terms of total number of products, we now have 45 16-bit microcontrollers in production, 30 more products are sampling, out of which 20 will be in production by the end of the June quarter. Based on the momentum with customers we see, we expect 16-bit microcontrollers to be up another 20% sequentially in the June quarter. Serial E-squared memory products net sales were flat, sequentially. We are continuing to focus on serving our strategic customers whatever we can attach Serial E-squared to a microcontroller or an analog product. Pricing remained essentially flat quarter over quarter. Analog products, analog products showed very strong performance and were up another 17.4% sequentially, and up 52% over a year ago quarter. We believe that we outgrew nearly all of our analog competitors and are gaining market share in analog. The design momentum on analog products has been very strong and we see continuing growth ahead. We expect a growth of about 7% in the analog order, we expect a growth of about 7% in analog in the June quarter, sequentially. Now our distribution strategy. Some investors and analysts have asked us questions about our distribution strategy. Microchip has two kinds of distributors: global distributors and regional distributors. Regional distributors are mainly in Asia and Europe, and they usually do not carry competing lines. These regional distributors have significant dedicated technical resources for Microchip, and they are demand creators. We are continuing to add regional distributors in Europe and Asia. The second kind of distributors is global distributors with a very broad line card. These distributors essentially carry most, if not all, of our competitors' products too. Therefore, the loyalty of these distributors is essentially split, and the technical resources are pulled into too many different directions. Microchip has not seen these global distributors being very efficient in creating demand. On the other hand, these global distributors are very efficient in fulfilling demand once the customer has designed in a product. So over the last two years, Microchip has steadily taken more and more control of its demand creation by adding a significant number of direct sales and technical application engineering resources. At the same time we have lowered the margin we paid to the global distributors since they largely provide fulfillment. This strategy has worked extremely well. Our continued success in 8-bit microcontrollers, and a very strong growth in 16-bit microcontrollers as well as in analog is essentially vindicating this strategy. We have also seen other semiconductor suppliers like Altera, Xilinx, Analog Devices and Linear Technology make similar moves. The large number of resources that Microchip has added to direct as well as regional distributors is creating record demand for our products. Fab capacity. I wanted to give you an update on Microchip's fab capacity since that's another question we often get from investors. We have been telling you that fab 2 and fab 4 combined, we have a total install the capacity of about $1.4 billion. Beyond that, we have additional clean room space available in fab 4 to take that number higher than $1.4 billion. But in the past, we have not dollarized that number. So we are prepared to do that today based on fresh analysis that we have completed. We believe that additional clean room space available in fab 4 will allow us to take the internal capacity to about $1.7 billion. Second, we have been silently building a Foundry strategy for a portion of our business, utilizing such technologies which are either not available in Microchip's sad or they have lower volume and hence not cost effective to develop them internally. We will not be giving any details of what we plan to build outside due to competitive reasons, however, because of very proprietary and value-added products being developed at Foundry's, we do not see any of impact of that on our gross margin percentage. We currently see the Foundry business becoming about 10% of Microchip's sales over the next five years from the low single digits, it is today. So combining these two factors together, we believe that we have fab capacity to take Microchip from today's run rates to about $1.9 billion. Therefore, we are confident for low CapEx and very high free cash flows well into the future. Now I shall discuss our guidance for the June 2006 quarter. As we look at the June quarter, we took several factors into account. First we saw only a muted effect of lunar New Year driven seasonality in the March quarter. Therefore, June quarter will grow from the base of a pretty strong March quarter. Second, we achieved record bookings in the March quarter with a book to bill ratio of 1.07. The bookings so far in April are extremely strong and at a record pace. Third, we have strong momentum in all of our strategic product lines of 8-bit flash microcontrollers, 16-bit microcontrollers, and analog. So taking all of these factors into account and after checking expectations from our direct as well as distributor channels, we expect net sales in the June quarter to be up 5% to 6% sequentially. Gross margins are expected to be about 60.3% to 60.5% this quarter. I remind investors that our long-term growth margin guidance is 62% and we're inching up towards that goal with about 25 basis points improvement every quarter. This gross margin guidance is without the effect of equity compensation expense. Earnings per share, again without equity compensation expense are expected to be about $0.37 , $0.02 higher than the last quarter. The earnings per share with equity compensation expense for the June quarter should be about $0.35, so a total of $0.02 impact from the equity compensation. We expect to build approximately $120 million of net cash flow before payment of $45.8 million of dividends just announced today and we look forward to sharing this cash with investors with another healthy increase in dividends in the next quarter. I also wanted to briefly talk about the lead times. Despite exceptionally strong bookings, our lead times remain in the three-to-six weeks range, which are the shortest amongst our competitors. We are monitoring inventories at Microchip, as well as at our distributors, very carefully. So far, last quarter our inventories went down by five days while our distributors' inventories were flat in number of days. Therefore, we do not see any overbuilding of inventories at Microchip or in our distribution channel. Finally, I have been telling investors and analysts about my new book for about a year now. The book titled Driving Excellence was finally released two weeks ago. I hope that all long-term investors and analysts will read the book. It gets you inside Microchip to look at how the aggregate system of Microchip was built and implemented. It reveals the culture of Microchip technology, which we believe is behind the exceptional performance of Microchip for so many years. The advantages of our products, technologies and manufacturing can be more easily understood, the company built on just those platforms can be [seen] over time. The real gap in Microchip performance is credited to our exceptional culture which remains the backbone of everything we do. So please read the book, and I would love your feedback. You may also visit to www.drivingexcellence.biz, which is a website created for the book for some additional information about the book. There's also a link if you want to buy the book. Let me summarize a few key points. Our net sales are expected to be up 5% to 6% sequentially. Now without the equity compensation expense, our gross margins are expected to be about 60.3% to 60.5%. Operating expenses to be about 24% of sales; operating profit to be about 36.3% to 36.6%. Earnings per share are expected to be about $0.37. Finally with continued strong growth of our 8-bit microcontrollers and very strong growth of our 16-bit microcontrollers, as well as analog products, we hope that we have put our investors' and analysts' concerns to rest regarding the continued top line growth of Microchip. With that, Justin, would you please poll for questions?

Thank you. (Operator Instructions) We'll take the first question from Adam Parker with Sanford Bernstein.

Yes, hi. Just a couple of quick ones. Steve, in round numbers you're getting pretty close to a $1 billion revenue company. If I just think out over the future here, if I think about your next billion in revenue, when you're a $2 billion revenue company maybe a few years down the line, how much of that new $1 billion in revenue is going to come from 8-bit maybe versus 16-bit or analog? Can you just help us conceptualize that long-term, please?

Well, Adam, as you know, we continued to not want to break out numbers for 8-bit and 16-bit either now or really in the future. None of our competition breaks it out and the information will help our competitors, I believe, more than it will help our investors.

Well, I guess my question, Steve, is more about the strength in the 8-bit business, and maybe I'll ask it a different way, do you have any competition at all in flash-based 8-bit controllers? If so who is it? And when you lose a socket, if that happens, when you compete for it, why do you lose one?

Well, we have very substantial competition in 8-bit flash microcontrollers. We compete with Freescale, Phillips, Atmel, Renaissance, NEC, some of the really smaller competitors, like Silicon Lab, a very little piece of it, but they're there, and with Cypress, So yes, there are a large number of competitors that make 8-bit Flash microcontrollers. However, if you look at this Gartner Dataquest report, in the top seven manufacturers or so, only two were up last year, ST Microelectronics was up about only 3%; everybody else was down quite significantly, while Microchip was the only company that was up. So we're really gaining market share over the entire market by a huge, more than more than 10 percentage point difference between the growth or decline of the market, which is the growth of Microchip. If you want to look at that report, it's very, very substantial. In dollar terms, the larger portion of the growth continues to come out of the 8-bit microcontrollers, and that is a market relatively most misunderstood and investors continue to not see the potential in it because the overall market either is not growing or is declining, but Microchip has grown in 8-bit microcontrollers, very, very healthy every single year with the exception of one year, which was a calendar year 2001.

When you compete for something, can you share at all what happens when you lose or is that not even relevant?

Well, you know, we have approximately what 14% of the market in dollars, about 14-15% of the market.

That's for overall 8-bit., not for the Flash-based side, right?

Flash, if you look at the flash-based , I believe we have maintained for many years about one-third of the market. I think we are about 33% to 35% of the overall market, again in round numbers Again, I don't have that in front of me. So two-thirds of the market is owned by somebody else; either a customer has a legacy with somebody else's architecture, or somebody else beat us on some price or our salesman isn't present in that account. Or it's in Japan where Microchip has relatively small share, or it's a particular product, the density, the packaging, or something we do not have. It's a competitive market, but we continue to do, very, very well.

Would you say that legacy is the biggest reason? Or if you were thinking about legacy price, or your salesforce for Japan; that was very useful. Can you break down the biggest reason you might, you might lose out or?

No, there's really no way to put percentages on what's the biggest reason you lose out. It's different by customer, it's different by geography, it's different by account. I know more, but I wouldn't want to share it with everybody because then the competition knows the same thing.

It seems like in that in this quarter you really grew more rapidly than the peers in a lot of the previous quarters. I thought maybe if you would help, was there anything in particular that was driving, market-wise any new applications or anything that was driving the particularly strong performance versus the peers this quarter? It clearly looks like we'll continue in the June quarter given the guidance.

Our business is always very broad-based. We're really never dependent on a single customer, Our business grew about 5.3% last quarter and our largest customer continues to be 2% or so of our business, and probably the largest customer shrunk. So the growth continues to happen from 50,000 accounts we serve worldwide. It's very, very broad-based. By customers, by applications, by geographies, by channels. It is really large scale, huge customer business and it can't really be analyzed regarding where the growth came from. The best we can do is really give you the growth by region like U.S., Europe, and Asia, and by product line, which we have kind of done that, broken out 16-bit percentage growth which was whopping 46% and also very strong growth in analog to give the investors a comfort that we're really gaining a very, very strong beachhead in the two very strategic markets for Microchip for future growth; one being the 16-bit microcontroller, the second being the analog. The 8-bit microcontrollers are continuing to grow actually providing the largest dollar growth still, quarter over quarter.

All right. That makes sense. Gordon, are you going to provide the breakdown on COGS, R&D and SG&A for the FAS 123?

Not in guidance. We will obviously when we report earnings. But given that we've indicated at 7% to 8%. This quarter there won't be any impact on COGS because it takes a quarter or two for that to be reflected in the P&L, so the charge in the June quarter will all be in operating expenses.

Okay. Thanks guys for you time. Thank you.

The next question comes from Michael Masdea of Credit Suisse.

You said flat overall distributor inventories roughly, but I think your deferred income was up 17%. Is that just speaking to the strength of the distribution channel in the first quarter?

Yes, absolutely. Distribution did extremely well in the March quarter, and needed to continue to grow their inventories to remain flat, so that's exactly what you're seeing in the deferred margin line, Michael.

And that's still looking pretty healthy into 2Q from what you're saying if I read it correctly?

Yes, we would expect inventories to stay in this range.

Got it. Can you give us more color on that OpEx piece or just how this stock compensation is going to impact some of your profit metrics or your spending metrics for the second quarter? Any ballpark?

Well, if you look at an overall $0.02 impact, you know, really, none in the COGS line. So all that impact is on R&D and SG&A. I think you could probably blow back pretty easily.

But over time it's going to be more on the line, correct? And we can start to see some impact there. I guess what I'm trying to say is how much can should we reset our long-term thinking based on stock compensation, if at all, for your model?

7% to 8% on the net line, earnings per share. . obviously, we'll get some color over the next couple of quarters as it's fully deployed. It takes a few quarters for that to be fully represented.

Okay. That's fair enough. This is the last question. You've done a great job on the capital intensity side over the last couple of years, but walk us through, or help me understand a little bit , as you get into analog, more into of some of this direct sales as you've been doing for the last couple of years, and maybe even more into 16-bit, are there any dynamics that can tweak the capital intensity?

It really has no impact on the capital intensity. Analog business essentially shares the same processes like we have in the microcontrollers, same thing back in assembly and test, so we really have not added any business that cannot grow in our existing infrastructure.

That includes human capital too or personnel and when you're adding more FAEs, et cetera?

Well, the human capital is being supported by significant growth in the top line, so it's contained within the operating expense. We are adding people every quarter, but, the operating expenses are staying in that range 24%.

Great and this is the last one and I'll get off. You said it's not really head-to-head competition but you see it out there; what is your kind of long-term strategic view on doing more integration into your microcontrollers and potentially getting more into programmable analog and some of the other things that product that Cypress is going after?

Well, you know, we have those products in our portfolio. We have over 330 or 340 total microcontrollers products, there's programmable analog products in there. We just don't break it out and make a big deal about it. Those products we've had for about eight years.

Got it, thanks a lot, guys.

Moving on to Mark Lipacis with Prudential.

Thanks. Two questions, if I may. The lead times you had mentioned are averaging between three and six weeks out now. Could you just remind us if that was a change over last quarter of at this time?

There's no change from last quarter at this time.

Okay. Fair enough. Steve, looking into the June quarter, could you tell us how does this year, looking into the June quarter of this year, how does it compare to over the last two years? Does this feel more like 2004 or more like 2005?

I don't remember 2004, going too far back. I'm sorry.

Okay, thank you very much.

Next question comes from Chris Danely with JP Morgan.

Thanks, guys, just to follow-up on Mark's question. Sounds like the lead times are stable, but a little bit extended. Is it your intent to bring those down, or do you think they'll just stay here for the rest of the year?

Well, I think, you know, three to six weeks are the shortest lead time in the industry right now. Many of our competitors' lead times are 50% to 100% higher. We have very, very strong bookings; very, very strong momentum in essentially all of our product lines and I think it will be tough to bring these lead times shorter. They don't really come in less than three weeks, that's just the minimum lead time gets if you've got it under-stocked; but even if the product is available in stock, we essentially quote about three weeks or so, which takes time to finish our market in the appropriate quantity and design that the customer wants and ship it to wherever in the world they want the product. So the minimum lead time you always quote is about three weeks, so three to six weeks is really as short as it gets.

Just to talk about the Foundry business a little bit. Are these going to be parts you guys do not have out right now? Could they be non-microcontrollers or non-analog products?

We are not going to reveal anything further on the Foundry business. What I could say is none of those products are the products where we make inside of Microchip and we're going to farm it out. So these are all additional products which will expand our servable market and help our business grow in all different product lines.

But are they going to be microcontrollers or analog, or could they be something different?

All different product lines including those, yes.

Just to follow up on that, you said it's not going to affect gross margins, but do you think that this could keep your capital intensity declining over time, or do you expect the same kind of capital intensity with the foundry business?

Well, in five years it becomes 10% of the business, so it's very low-single digits today, so that's a fairly small impact. You're talking about 1.5% to 2% of the business going to the Foundry, which doesn't have a huge impact on capital intensity. Our capital intensity is low for the next several years, because internal capacity can go from the $1 billion that we are today to $1.7 billion and then the outside fab gives you another last couple of hundred million dollars to $1.9 billion. So it's a combination of all that.

Steve, can you just explain the decision to go and start doing ROM-based microcontrollers, and also are these gross margins the same as the other programmable micros?

We have done a few ROM-based microcontrollers, going back even many, many years. Even though essentially high 90% of our business is field programmable microcontrollers, the overall field programmable microcontrollers are approximately 40% of total microcontrollers. So, 60% of the business continues to be ROM-based microcontrollers, and our business is growing continuously by showing customers how to design with flash and design away ROM. However, that's a significant additional available market, and we'll continue to be there for years and years and years. Although the market is shrinking. So having really negligible entry into that market, if you can take a piece of that market while the market is converting to flash, I think it helps Microchip keep it growth rate higher, and then gives customers a solution in certain cases where they do want a ROM-based microcontrollers in our architecture. See what happens many times is customer would say well, I love your product, I love your architecture, it's a great product, I like your delivery performance, your cost, and everything else; we have done business with you on our Pic-based microcontrollers, but I have these applications where I would rather use the ROM. And if the ROM isn't available in this architecture, then they will go off and design somebody else. So it's basically allows us to incrementally capture a customer whose heart we already captured with our architecture.

Are they pin for pin compatible with other companies' microcontrollers?

Absolutely pin for pin compatible, yes.

Are the margins a little bit lower than your other products?

Yes, they're pin for pin compatible with our microcontrollers, not with the competition's.

Got it. And are the margins the same as your other products?

I don't know, micros would be similar, yes. We don't break it out on individual products, but it's not going to be enough of a business to really make any kind of difference.

Okay, thanks guys.

Next question comes from Simona Jankowski with Goldman Sachs.

Hi, thanks for taking my call. I just have a quick near-term question, first, and then a longer-term question. On near term it looks like your analog and 16-bit business has accelerated pretty significantly since your March 6th update. I just wanted to see if you have a sense of what drove that acceleration?

Well, there are is not one specific customer or something; we have a large number of customers growing and the business has done very well. In the analog business it was up about 17%, 18% last quarter, so I'm not sure it has necessarily accelerated. We did extremely well in December quarter, and this quarter was up again 17.4%. The 16-bit microcontrollers was up I think about 18% or 19% in December quarter and is up significantly more this quarter because there, the product went to production about two years ago, so there are thousands of customers with development tool in hand are designing from it, and you can't accurately predict within a quarter when the customer is going to finish design and start buying volume production. So there, clearly, more of the customers have gone to production and that business has accelerated, but for the analog, it went up 52% year-over-year and last quarter up 17%, so it really all wasn't driven by last quarter alone.

Okay and then just for a longer term, more industry trends kind of question. We've been noticing some interesting new introductions in the 32-bit arena which is in a very, very low cost; partly seems like they've been enabled by a new core by arm. I've noticed also that you've filed a lawsuit against one of the start-ups which just launched one of these $1, 32-bit microcontrollers.. And just wanted to get your perspective on, is this ever going to be a real competitive threat kind of down the line for the 8-bit space, or what would be the target opportunity for these kinds of products?

Well, you know, our 16-bit architecture that we're doing extremely well on, significant growth, 46% growth sequentially last quarter, essentially beats the performance of an arm. So , we're winning designs against arm every day today on a performance basis, these products are a hybrid architecture. They are 24-bit internally, 16-bit outward, just like for years and years our 8-bit controllers have won designs against products like AT12 and other 16-bit products. Our 16-bit microcontrollers handily beats many of the 32-bit microcontrollers out there. So we are basically duplicating the strategy on 16-bit which we have very, very successfully deployed for many, many years on 8-bit. And I expect over the next 10 years to really see similar comments for years and years after we continue to win, what are you going to do on 32-bit? how this product will play? Essentially we have competitive products with 32-bit out in the market today and winning against 32-bit designs every day.

Just as a quick follow-up to that. Do you think the price point though is something that has kind of changed compared to the 32-bit micros we saw years ago? Seeing them now in the $1 to $2 range do you think that's something that's a little bit new in the industry?

Well, you could similarly see over many years people, you know introducing 16-bit microcontrollers products as much as five years ago at price points where they said it's going to wipe out 8-bit microcontrollers and many have forecasted those demises and yet our 8-bit microcontrollers business continues to do records. It's very easy to really have a PR game on a price point of, $1, or $2 or whatever. What's on that micro, what are the peripherals, what is the architecture, what is the performance, how good are the development tools, can it win the engineers' hearts who design that product? So as we engage with the customers, we continue to beat all of these architectures, and the actual cost of ownership in customers' design is significantly, significantly higher than really what is advertised for the PR purposes, generating the kind of questions you're asking. I think we have a lot of experience having seen that in 8 to 16 and now 16 to 32.

Got it, thanks a lot, Steve.

This question comes from Romit Shah with Lehman Brothers.

Thanks and congratulations on publishing the new book. Steve, one of the smaller players in the analog space mentioned this afternoon that, sell-through is good, but just the orders could be flat to even down in Q2 as these guys are now getting back to more desired levels of inventory. You indicated that inventory levels at distributors are low, but are you encouraging them to carry even more just given the strength in your product lines?

No, we are not encouraging distributors to carry more inventory, we have absolutely zero incentive to push distributors to carry more inventory. The inventory in our hand can be shipped to any customer or any distributor worldwide within moment's notice. Inventory sitting on a distributor's shelf can only be used for that distributor customer only in that geography. So therefore, the inventory sitting in Microchip's warehouse in Thailand is substantially more efficient in being the customers worldwide. So therefore we don't really expect the distributor inventory to rise. Our distributor inventories are about two months. The lowest we've ever been is about 1.9. So they're sitting at a very, very low end and the highest number many years ago was close to three-and-a-half months, so we actually like low distributor inventory. We do not take any shipment to distributor worldwide as revenue. Therefore, we have no incentive to really push inventory into distribution unless it's going to get sold out to the end customer.

Okay. My follow-up is on incremental gross margin. I understand that the changes will not be linear, but the decline from the mid-70% range to the high 60s last quarter. Could you just comment on the factors that drove the decrease in margin? When I think about your guidance, it sounds like you expect gross margins to be in that high 60 range. Can you comment on some of the factors, i.e. mix, or perhaps higher depreciation?

First off, what decline, Romit?

I think he's doing and incremental gross margin calculation as a incremental gross margin dollars divided by incremental gross margin revenues.

That's correct.

We just don't look at the business that way. There is too many different moving parts, and as we go from 60% gross margin to a long-term 62% gross margin, you basically get to the point where that incremental goes to equal to the total gross margin, so you have asymptotically approach that number. Otherwise your gross margin will continue to go higher and higher into the stratosphere which is probably not healthy for the business.

Thank you.

You're welcome.

Moving on to Shailesh Jaitly with Nomura Securities

If you could help me understand the Asian part of the business. Is it more Company-specific, or are competitors also seeing similar declines?

The Asia part of the business, if you look at the last several quarters, June 2005, September 2005, December 2005;, Asian business was up very, very strongly. The March quarter in Asia has a large number of holidays because of Chinese lunar New Year. So our business is down in the March quarter every year seasonally. Asia business will be up very, very healthy in the June quarter. I haven't seen competitors' numbers and certainly even if I read the numbers on a top line basis, I do not dive into how the businesses are doing geographically, I think that's probably your analyst job, but I'm sure they have similar impact on the lunar New Year.

Understand. So pretty much seasonal.

Yes, absolutely.

Sure. Thank you.

This question comes from Chris Caso with Friedman, Billings Ramsey.

Yes, thanks. I wonder if I could just ask about the analog business again a bit. Could you talk a bit about perhaps what you think your competitive advantage is with that business? Obviously there's a lot of capable competitors, what is driving the gains that you guys seeing right now? A high-level look at the business.

Well, I think I can only answer it very, very generically without giving any of the secret sauce away. You know, Microchip has said for many years now that our microcontrollers business is a highly analogous, in fact, well in excess of 80% of our microcontrollers business has very, very strong analog on it. The microcontrollers products are loaded with A2D converters, power management, supervisors, reference voltages, and stuff like that. Microcontrollers is a real-time control unit, while the real world is analog, so whether it's temperature, pressure, motion, motor; anything microcontroller is trying to control, it requires sensing and taking in analog signal, converting it to digital, and then doing whatever you want to do with a microcontroller and then doing something on the output side which is again analog. So Microchip has had substantial analog expertise for a very large number of years. Several years ago we recognized that despite having a large amount of analog in a microcontroller -- by the way, our analog microcontrollers, microcontrollers which have analog in them, is over a $600 million business at Microchip. So we have a large amount of analog expertise. In fact we more analog expertise in our microcontrollers group than in almost outside of it. But over the last few years, we sense that no matter how much analog we embed inside the microcontrollers, there's always a strong presence of standalone analog products around our microcontrollers. And that's driven by the fact that, analogs are so ubiquitous, they're so fragmented, and not every customer wants to use every analog feature inside the microcontrollers. Some want to use it outside, sometimes there are advantages of using the product outside. They can be faster, or lower powered, or they can be multiplex outside while you can multiplex it from inside the Microcontrollers. So seeing all of those trends, Microchip decided to be in the analog business, and we have taken a large amount of analog capability inside the microcontrollers, packaged it standalone, and we have developed large amount of additional analog capability and many of them are best products of its kind. Every year we win a number of awards from various magazines from leadership regarding best this and best that. We've got the best low-powered options in the industry, we've got great converters, and lots of other different products. And we've been telling investors for a number of years that our analog business essentially will compete with a head-on competition with all the other major analog competitors: Maxin, Nenia, ADI, TI and today we're standing up to everybody in winning designs. We got started a few years ago by trying to attach analog to our microcontrollers business because that was the easiest, the customers had name recognition, they understood our products and our initial success came from selling analog around our microcontrollers. But then we went to the phase 2, which was selling analog to any microcontrollers customer whether they used our microcontrollers or they used our competitors' microcontrollers. If we lose a microcontrollers design to free scale, we want to still be able to win analog. So that was very successful. The phase 3 in the last two years has been basically selling analog to anybody, whether they use a microcontrollers or DSP, or SPG or microprocessor, or any kind of application, and that's the phase we are in. Phase 1, 2, 3 are all applicable, and growth has accelerated. Analog business was up 52% last quarter over the same quarter a year ago. That's the broader answer rather than trying to pin it down what product line is growing, which we are always uncomfortable in signaling.

Well, I guess it was the December quarter where you started to show double-digit growth in the analog business, and I realize that you've been exiting the handset portion of the analog business for some time. Is there something in particular you could point in the December quarter that would explain the out-performance there, and is it something that you expect to continue over the next several quarters?

Well, we certainly expect the analog business to continue to go very strongly. Remember the handset decline slide, Microchip is a company that doesn't like the handset business. It's a huge volume, low-margin, fast-changing, so we managed the decline of the handset part of the business over the last four to five years, and now it's mid single-digit of our total analog business so it really can't hurt any more. As that happened, you're seeing the real growth of the non-handset business show up, and we think that growth should continue. Over time, you'll have great quarters and good quarters and sometime even not so good quarters. So I can't tell you it's going to grow 17% every quarter, but it's growing more than the overall market growth of Microchip. And I'm guiding 7% growth to the next quarter, I hope to find it conservative, but I can't guarantee that.

Just final question with regard to the incremental gross margins for the analog business, where your corporate gross margins are right now, the analog business is pretty much in line with that?

I would not want to comment on that. Relative to -- corporate gross margins are 60, that's pretty good -- and analog gross margins are extremely good, but I don't want to tell my competitors exactly what those are.

Extremely good meaning a little bit higher perhaps?

I just want to leave it at that. Again, I would like to tell the investors and analysts that, to maintain a, 60% gross margin business in a dominant microcontroller-based company which is 80% of our business; just in the last hour you heard the results on Atnal, and you heard the results on Freescale last week or whenever. I want to create a performance that is so far above our similar competitors which requires a level of secrecy that just I'm not willing to break. I cannot afford to take any risk on anybody knowing anything they should not know.

Okay. Fair enough, thank you.

Thanks.

(Operator Instructions) Next question comes from Jeff Rosenberg with William Blair.

Hi, when you look at the competitors you were talking about being down in the Gartner numbers, are they down on a unit basis or is one of the differences a greater decline than ASP versus what you see?

Let me have Ganesh answer that question.

The report at this point really does not break down units in ASP, it's only in dollars, so it's the revenue that is reported by the individual companies.

I thought usually they give both. Well, the follow-up I was going to ask was on the 16-bit business, are you today only shipping DS Pic? Is that really what's in the market today, or have you started to ship and see revenue from the 24 series?

We have 24x products are in volume, but the contribution is quite small because they basically just went to production.

Okay. Would you characterize the initial acceptance of the 24 series any differently than what you saw in the early stages of DS Pic?

It's way too early to tell, way too early to tell. This is the first quarter.

I was thinking about it from a design perspective.

It's widely acceptable. Can you comment on this new survey that we just received?

It's a survey that's done by ED Times and it's done by a company called CMP. If you look at how they rate 16-bit microcontrollers in the different brands from different companies that come about, Microchip with its 16-bit microcontrollers is the highest rated by design engineers looking to go start new designs with a 16-bit microcontrollers. I don't remember the exact numbers; like 29% or so of the designers are looking to a Microchip solution. The strength of that is 16-bit is not a new market for us, 16-bit is a market we've served for the high-end 8-bit for many, many years. So many of the same customers who we have 8-bit with, are also looking at our 16-bit product line as a natural extension of what we've done already with them.

See, eight or 10 years ago when we were number 12 or number 10 in 8-bit microcontrollers and climbing, we used to get these kind of survey results, the top microcontrollers architecture pick that customers were looking at. It takes years to reach that position because many people are looking at it, but when the rubber meets the road, not everybody designs with it, you lose because of legacy or other things and product line or whatever. But every year our 8-bit microcontroller share grew and the history shows you the results. We're seeing exactly same kind of results coming out in server, in design wins. So, we're very confident that the path we have on 16-bit is a winner.

Okay. Thanks.

This question comes from Harsh Kumar with Morgan Keegan..

Hey, Steve. First of all, congratulations on a great quarter, great guidance. Most of my questions have been answered. This one question. Flash-based microcontrollers, where do you see them peaking out on a percentage of business? Just any clarity on that?

Percentage of our business or total conversion in the industry?

Your business.

Well, you know all growth of Microchip and microcontrollers is coming from flash-based. We are essentially not doing any products which are not flash-based with the exception of a few ROM-based we have done. In fact, the non-flash-based is declining, so I think I do not know what will be in five or ten years, just going to get high 70s and 80s. and maybe even higher than that.

Got it. Thank you that's helpful.

This question comes from Eric Gomberg with Thomas Weisel Partners.

Congratulations on a great fiscal year.

Thank you.

Since you don't provide much specificity other than the sequential growth on the 16-bit products, could you maybe give us some metrics or some ways of thinking about how things are coming along in fiscal '07 -- whether it's, you're at 324 customers today, what would be a signal that things have gone well over the course of this fiscal year, or what type of sequential growth would you anticipate throughout the year so that we know that things are progressing as you hope?

Well, I have given that number of customers every quarter in the past. So I have it on a graph. I'd certainly be happy to send it to you. We should continue to see additional number of customers, we have also given the development tools and we should continue to sell, large number of development tools and beyond that, we should be reporting strong sequential growth on 16-bit microcontrollers quarter over quarter. The fourth would be the number of products. We have 45 products in production today, 30 products we are sampling, and out of those 30, almost 20 will be in production by the end of this quarter. If you look at combination of those factors, I think that we are continuing to really make a significant headway in that market.

We guided to 20% potential growth in the June quarter here.

Right. Okay. Question on SG&A. Maybe it's because you did more internal demand creation, but SG&A actually had reverse leverages up fiscal year over fiscal year as a percent of revenue. Do you expect that to trend in line with revenue over the course of fiscal '07, or do you expect to get some SG&A leverage as revenue builds?

I think the key here is what Steve spoke to in terms of the model we have. It's obviously the approach we have taken with our distribution partners relative to wanting to have control of the design creation element of our business. It does mean that there needs to be an investment on our part in highly technical field sales engineers, as well as field application engineers, so we'll continue to invest in that area to grow our business. It's really very important that we're able to drive our customer designs to completion.

Eric, my numbers show fiscal year '05 operating expenses were at 24.1% of business, and fiscal '06 were at 24.2% of business.

I was just asking about SG&A, which was up.

Okay on SG&A clearly we're making significant investments in SG&A because of demand creation and that's why we're growing substantially much, more than the industry if you look at the Gartner report, in a market where everybody has been essentially not giving us the credit that 8-bit can grow. We just recorded numbers which were up, what in microcontrollers year over year?

19%.

19%. So those are industry beating results, and I think when you're as profitable as we are, 36.2% of sales as the operating profit, if you do not make the right investments to grow your top line, then I think shame on us.

I think that ends up being the metrics. I mean gross margins were up 275 basis points, and operating expenses are flat, but certainly more investment in the sales and marketing component of our business, not the G&A. It's in the sales and marketing area that drives our design penetration. So the end result is leverage at the operating profit year to year. You can't look for every component of the P&L to have leverage. You have to invest in the future, and part of what we're doing here is investing in the future growth of Microchip with the resources that we're putting in place.

That's fair and clearly it's been paying off. And one last thing, you said your competitors have 50-100% longer lead times right now. Just wondering to any extent if any you think that's benefiting you in terms of potentially bringing you new customers or folks out there who are having problems sourcing from their current supply base.

It does not help you at all in the current quarter because there is no microcontrollers customer who is using someone else's device can simply plug in our part. These are proprietary products with minimum six to nine months design cycle and sometimes longer; up to a year or two year in design prototyping and all of that. So we're getting no effect of somebody's longer lead times in terms of very short-term. But are you winning the customer's heart in terms of they're hurting from other people's lead time, so for the next design they're looking at Microchip, which would have an impact nine months out, a year out in their next design? Sure. That's how you, by doing a better job on a lot of different fronts, sometimes you win a design because a customer is upset with a competitor. Sometimes you win it because, you got a better price, sometimes you win it because you've got better features and sometimes you win it because the customer is now willing to break the legacy because they're very unhappy with the service of somebody else. Again, like I said, when you have these kind of results like we do in a market that's largely flat on 8-bit microcontrollers, we can grow 18-19%. Last quarter over quarter year ago, you need everything possible. And we're taking advantage of everything possible. We are succeeding in a lot of different dimensions.

Thank you.

Thank you.

Moving on to Tore Svanberg of Piper Jaffray.

Yes, good afternoon, great quarter. I know your development tools can sometimes be a pretty good correlation to the success of your business. Is there a way to maybe quantify that? Quantify that correlation?

There was a correlation many years ago, but if you look at the growth of the total cumulative development tools and you plot the revenue, it was a very good correlation. Now we have 432,000 developing tools in the market, and shipped over the last 13-14 years and around the world and we do not really know how many tools that were shipped several years ago are still being actively used or the tools have been thrown away or the Company went away or the engineer got laid off and the tools got junked out and buying more advanced tools. It's very, very difficult to get a good assessment when you do that correlation. It's just, from time to time, we have tried sampling a couple hundred customers who bought the tool in the last nine months to a year to see where they are in their design cycle to try to do some metrics on where the business is headed, where we are winning, where we are losing. Those metrics are confidential, but the correlation on the total tools is not very good. The tool sales also jumps up and down based on a number of products introduced, especially new architectures introduced, like we introduced a number of 16-bit architectures so a lot of people have to buy tools. It also runs by promotions, when we're doing seminars, tools should be very, very heavy in July, August, September quarter because we do a [Napster's] conference and nearly 1,000 customers come to that conference and a lot of people want to buy tools there. So, you know, it's really I can't do the correlation to the level of granularity in investors looking at which is plus or minus zero essentially.

That's fair. And also on your Foundry strategy going forward becoming 10% of sales, would that ratio also apply for both MCU and analog?

I think I answered that question before by saying we're going to be building a range of products, but we won't really identify for competitive reasons which products we will build because I think it's a question of for the competition, piecing it together. They get one information one day, what products you do, next time they hear, a sub product line, and next time they hear, technology, the next time they hear what the wafer costs at the Foundry might be, and we don't want them to get at really what we can make and what prices and what lithography.

Fair. And then last question. I think you mentioned for fiscal '07 the 8-bit business continuing to grow, driven by some new and emerging applications. Can you give us a couple of examples of those?

Well there're hundreds, I think, about nine months to a year ago we were reporting 46,000 customers; today we have over 50,000 customers, so our business is always thousands of additional customers. Hundreds of additional applications, microcontrollers, it is ubiquitous, it goes into everything. So I couldn't really pick out one or two because our business just is not killer application dependent.

That's fine. Thank you very much.

Next question comes from Sumit Dhanda with Banc of America Securities.

A couple of questions. Gordon, could you give us some sense of how distribution resales did in Q1, and how you think that might shape up; if you have any idea in calendar Q1 and calendar Q2?

Well as we indicated distribution performed extremely well in the fourth quarter. You can see that reflected in our deferred margin line where the inventory stocking was up to reflect the fact that sell-through was extremely strong. I think if you look at the geographic elements between Europe, Americas and Asia, you can get a reflection of what was happening from a distribution perspective. As well, Europe and Americas were very strong in the distribution area. We would expect more normal trends based on seasonality and geographical dispersion here in the June quarter from each of the geography.

Is it fair to say that your deferred income was up about 17% or so and distribution resales were perhaps not quite that number, but close to that and higher than the overall revenue growth you posted in the quarter? Or is that stretching it too far?

The distribution was higher than the overall revenue growth and that's why you're seeing the inventory levels increase to meet the demand that they're seeing.

Steve, in your prepared comments you mentioned growth in 8-bit controllers driven by new applications. Any particular applications you could highlight which are gaining significant traction in driving your growth?

The same question the gentleman asked before you. Did you hear that?

I apologize, I've been shuffling back between calls. I'm sorry if I missed that.

The answer I gave to that is there's no single application that drives the growth in our business, We have 50,000 customers, we serve thousands of applications, business is very, very broad-based. Microcontrollers is the most ubiquitous device you can find that goes into everything that has power applied to it and from our broad-based numbers we can't pin out any single application that's driving the growth.

Okay. Thank you very much.

You're welcome.

Moving on to Steve Maresca with UBS.

yes, this is Steve Maresca for Tom Thornhill. I have a couple of questions. First with regards to your investments in operating expenses, you talked about SG&A in those investments to take more customers direct. With regard to R&D, that's been held fairly flat. With regards to investing into new areas, do you see R&D potentially going up here as a percentage of sales over the next year or so?

It should not go up as a percentage of sales. We can grow in dollars by staying into a model and simply investing the incremental revenue we are creating.

Okay. So this $23 million, $24 million we should kind of see that you're going to maintain that level.

10% of sales.

As percentage of sales we should maintain it, but the number will go up in dollars as the revenue grows.

Second question is with regard to your memory business, you don't talk much about that. Should we think about that kind of as a single digit year-over-year growth business, or are you making investments there that perhaps you'll tell us more about?

Well, the memory business is now about 11% or so of our sales, something like 11% to 12% of our sales. And five years ago it was over 30% of our sales. So it is a less strategic component, it's not like we're going away from that business. It is an extremely profitable business, a very good business. However, memory is not as design-end driven business as analog and microcontrollers are. The margins are significantly higher on microcontrollers and analog than on the memory business, but it's a separate division, so it's not like it's competing for resources or anything else. It is introducing its own new products. In the last year it introduced several new products in the area of 512 E-square and megabit E-square-- in fact memory business last quarter was up year-over-year , it was up about what percentage?

I think 9%.

So it was up about 9%.

So it's about 12% to 13% of our business to-date.

Okay, thank you.

You're welcome.

This question comes from Mark Edelstone with Morgan Stanley.

Hi, Mark.

Hi, thanks a lot guys and nice job on the quarter.

Thanks.

You talked about basically expanding the clean room or building out the clean room of fab 4, I assume, to get to $1.7 billion of internal production. Can you talk about what the expected capital spending would be like to get that incremental $300 million of revenues?

The clean room was already built out when we bought it. It's just when we bought the facility, you know, it was not all equipped. It was substantially equipped, but there is a lot of additional clean room available in which we can install equipment. The equipment we have to install in it is not really in one day. It happens over the next five years as we go from here to completing the total build out, facilitization of the fab. Most of the equipment we can buy it really, the used equipment or whatever at a fairly discounted prices over time and the capital required should really fit into our budget of the kind of percentage of sales we are capitalize percentage of sales, which in current year is about 8%. So somewhere in that 8% to10% range we should be able to complete all of that. So there are no extra expenses needed from outside.

Understood. I guess I was just wondering what you thought the capital expense would be like to fill that equipment setup over time.

I probably wouldn't want to give that number, but I think I tell you with an 8% to 10% of sales will be the capital budget for the next several years, and in that is included that equipment probably the best I could do.

Maybe just to think about it differently. What do you see as your maintenance capital spending at this point, just to keep facilities operating at high levels?

It's tough to split it into those areas. Obviously we invest in new capacity, we invest in our assembly and test area, to complement what we already have, but certainly with our business we do have to look at existing equipment, and there's the need to refresh that on occasions, I don't know that I have got a metrics that I could pull out that would have a real historical view over a few years that I would be comfortable with.

Okay, thanks a lot. Just one other question with the end of the fiscal year, Gordon, have you guys taken a cut at revenues by end market?

No, we haven't had a chance to do that as yet.

Okay. Thanks a lot, guys.

Thanks, Mark.

Moving on to Craig Ellis from Citigroup.

Hi this is Terrence Whalen for Craig Ellis with a quick question related to the distribution versions OEM. Assuming distribution did grow in line with the deferred line of about 17%, it looks like OEM then declined about 15% in revenues. Is that decline of OEM, is that reflective of inventories being worked down or actual patterns of demand?

Those numbers are not accurate. Both OEM and distribution grew. There's some relationship in terms of the deferred margin, which doesn't always cover exactly the unit growth in distribution, so I think that there's data which is peeling the onion two or three levels that you're not able to see from a gross balance sheet perspective, but I can confirm that both OEM and distribution business grew sequentially in the March quarter.

Okay. Great, that's helpful. I was looking just at the deferred income line and assuming that revenue would grow.

I understand.

My second question is related to distribution inventory but in a different way. You mentioned your distribution is going more direct. Does that mean that we can assume a structurally different level of inventory, maybe say below two months going forward in the channel? Thanks.

What we said is, we are taking more control of demand creation, but I also said in my prepared remarks that a lot of the shipment of that business, fulfillment of that business continues to but through distributors. So we're really not thinking here that there's going to be a dramatic change of OEM versus distribution mix, but the large portion of distributorships, actually a demand is created by Microchip. It's Microchip's engineers and sales guys that really create that demand and then the customer wants to use the distributor as a supply-chain channel.

Great, thanks very much.

That does conclude the question-and-answer session, I'll now turn the conference back over to you.

Thanks very much, Justin. I appreciate it. I appreciate everyone's time on the call this afternoon. I appreciate the questions and the opportunity to respond to those. Steve and I will be around able to answer additional questions as required, thanks, everyone. Take care.

Thank you. That does conclude today's conference call. Thank you for your participation.