FormFactor Inc (FORM) 2005 Q1 法說會逐字稿

Good afternoon. My name is Angela and I will be your conference facilitator today. At this time, I would like to welcome everyone to FormFactor's first-quarter 2005 earnings conference call. As a reminder, this call is being recorded and will be available for replay on the investor section of FormFactor's website, www.formfactor.com. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question-and-answer period. (Operator Instructions). At this time, I would like to turn the conference over to Erica Mannion, Investor Relations for FormFactor. Please go ahead, ma'am.

Thank you. Good afternoon and thank you for joining FormFactor's first-quarter 2005 earnings conference call. With me on today's call are Igor Khandros, Chief Executive Officer; Joe Bronson, President; Jens Meyerhoff, Chief Operating Officer; and Ron Foster, Chief Financial Officer. Today, Igor will provide you with a summary of our first-quarter performance, a review of our marketplace and our overall corporate strategy and highlight our recent product announcements. Joe will then review our corporate priorities and the significant changes to our markets since our last call, and Ron will review our financials and provide our second-quarter guidance. Please note that details of the Company's revenues and bookings by market segment, as well as the regional breakdown of its revenues, can be obtained from the investor section of FormFactor's Web site at www.formfactor.com.

I would like to take a moment to mention that during the second quarter of 2005, the Company will be presenting at the following conferences -- CIBC's annual communications and technology conference in New York on May 10, Piper Jaffray's technology conference in New York on May 11, and Citigroup's semiconductor conference in Monterey, California in early June. As the dates of these conferences approach, we may make additional announcements. Finally, before I hand the call over to Igor, I will review our Safe Harbor.

During the course of this conference call, we may make projections within the meaning of the federal securities laws, including statements regarding FormFactor's growth and financial performance as well as our strategic and operational plan. These forward-looking statements are based on current information and expectations and are inherently subject to change. Actual results may differ materially and adversely to those in our forward-looking statements due to various factors, including but not limited to the rate at which customers adopt the Company's newly released architectures, the technologies and products, the extent to which chip manufacturers modify announced capital equipment expenditures and device roadmaps, and the Company's ability to efficiently complete the transition into its new manufacturing facility. Please refer to the Company's recent filings on Form 10-K and 10-Q for more detailed discussions of the relevant risks and uncertainties.

FormFactor undertakes no obligation to review or update any forward-looking statements or update the reasons actual results could differ materially from those anticipated in forward-looking statements.

I would now like to turn the call over to Igor Khandros. Igor?

Thank you, Erica. Welcome, everyone, and thank you for joining us today. As many of you know, we are in the midst of an extraordinary time in FormFactor's evolution. We have grown over 80% in 2004. During the first quarter of 2005, we continued to increase our overall capacity at our existing facilities and we simultaneously proceeded with customer qualifications at our new manufacturing facility, in line with our stated plan. Managing such transitions is a challenge for any company, and we believe the FormFactor team is performing well, as evidenced by our strong quarterly results of 51 million in revenue and 10.5% increase over the fourth quarter of 2004.

The overall marketplace for FormFactor is healthy and our largest market, DRAM, continues to be robust. We believe we have created a resilient business model from which we've benefited every stage of the DRAM business cycle. When the business is growing rapidly, as big growth drives capacity expansion, customers who issue (ph) capacity for various memory applications turn to us to enable a turnkey test strategy for their manufacturing. When the economics of the business become more challenging and capital preservation is paramount, customers must seek ways to maintain margins by lowering the cost of sales and moving to next-generation products. Today, we continue to move into a design-driven part of the cycle where customers are pursuing increased design activities as they look for ways to increase their revenue per wafer.

On live semiconductor capital equipment companies (ph), FormFactor's valuable position increases during this stage of the cycle as our cost and product are required for each and every new design. This uniquely positions us in the test industry to sell products to customers at all phases of the industry cycle from new designs to rapid capacity expansion to aggressive cost reduction.

In order to maintain and extend our leadership position in 2004, we placed additional focus on R&D and product innovation. These efforts attained dividends in the form of exciting new technologies and architectures, which fuel our product pipeline. As our enhanced growth of capabilities and increased capacity fall in line with our new factory, we will further leverage this effort. Our mission has been to drive higher testing parallelism, which will eventually culminate in one touchdown testing, testing all done (ph) on a fully populated wafer in a single contact.

I'm very excited about the announcement that FormFactor made yesterday regarding our shipment of alpha to touchdown proof guards for 200-millimeter Flash wafer testing in early evaluating customers (ph). This product is capable of testing 792 Flash devices and parallel (ph). Most significantly, it is based on our brand-new Harmony architecture, which we believe will eventually enable our customers to get 300-millimeter wafers and beyond in a single touchdown. We will leverage this new architecture as the basis for our two touchdown and one touchdown wafer testing product, which we plan to introduce over the next few years. This new evolutionary architecture will be the key building block of our future DRAM, Flash, wafer-level burn-in, and high-frequency product platform.

Our ongoing R&D efforts have resulted in industry-leading products and technologies, and we are protecting our investments through our strong IP portfolio. As a key part of our strategy, we have and will continue to defend our IP. To this end, we received another favorable ruling this month in the Korean intellectual property office, which has now upheld the validity of all four patents. We're starting in our current Korean infringement litigation.

I will now turn the call over to our President, Joe Bronson, who will cover the quarter in more detail.

Thanks, Igor. FormFactor had a strong first quarter. As revenue grew 10.5% sequentially to 51 million, exceeding our projections of 37 to 38.5 million, base margins also expanded to 43.4 and net income grew to 4.9 million. We were very pleased with the Company's performance as we continue to focus on improving our manufacturing capacity, on-time delivery, new product introductions and the startup of our new Livermore factory.

The transition to our new manufacturing facility is on track and its customer product qualifications are proceeding. Customers have been impressed with the new facility and its capabilities. We expect to exit the quarter at planned capacity levels and have changed the production capabilities we discussed on our last earnings call. As Igor mentioned, building up a new factory while ramping capacity with our existing facility is a major focus, and we are pleased with the team's execution.

Semiconductor demand continued relatively flat in the first quarter, with a decline in equipment purchases. Current pricing declines in DRAM and related products were evident (ph) across the board. At the same time, demand-based bid growth continued in the DRAM market and companies tooled the transition to DDR2 and to 7 and 10 nanometer. The buildout of 300 millimeter capacity remains on track with a substantial amount of equipment already ordered or in installation. These transitions, coupled with customers' increased design activity and the margin pressure, contributed to our strong quarterly results this quarter. We remain well-positioned in our market. Based on our preliminary reports, we believe that we have gained market share in the overall test market and advanced probe card market. Our unique strategy positions us to benefit at all stages of the market -- technology transitions, architectural changes and capacity expansion. Not only is there a high percent of DRAM devices in the tooling cycle, (technical difficulty) DDR2 devices, but only we can test cost-effectively.

And as the economics of the business become more challenging to our customers, we become increasingly mission-critical in terms of our ability to provide lower costs of test, which is a key strategic advantage. And to revision (ph), the DDR2 remains strong as customers continue to ramp their production capacity. First-quarter DDR2 shipments rose substantially, while bookings fell slightly. We saw DDR1 reaccelerate. As some of the 90 nanometer learning (ph) is taking place at the DDR1 level. However, we believe that big growth in DDR2 is accelerating and bid parity will be reached in the third quarter. We continue to expect the size of our DDR2 market to be twice that of the DDR1 market at 300 million.

Technology transitions are among the most powerful tooling cycles in our business. As the transition to 90 nanometer continues to gain momentum, we are positioned to benefit from this evolution as currently 93% of our DRAM revenue is derived from 110-nanometer-and-below products. This is reflected in our shipments for 110 and 90 nanometer designed in the quarter, which grew 31% sequentially. 90 nanometer (technical difficulty) the tooling cycle that was expensed throughout the year.

Capacity expansion, another one of the global drivers, continued in DRAM and Flash, as seen in the buildout of foreign (ph) capacity. Our shipments in bookings of 300 millimeter grew significantly in the quarter. Based on our strong market share, we expect 300 millimeter to continue to be an area of expansion for the Company. As companies build new 300 millimeter fabs, we will require length of test capacity through 2005. It is important to note that we are only a third of the way through this major transition.

In global RAM, we're showing the success of our expansion into other markets, particularly consumer application. Earnings and bookings in mobile RAM increased substantially in this quarter. This was due largely to improved consumer application demand in Japan in mobile RAM becoming a greater percentage of the overall DRAM market, as it currently commands a pricing premium over commodity DRAM. As the DRAM market continues to bifurcate, we expect to see mobile RAM as a good example of how dueling (ph) production is shifting to higher revenue-per-wafer opportunities. Long-term, we see mobile RAM as an area of growth for FormFactor.

The environment for Flash remained virtually unchanged from the fourth quarter, especially as demand for NOR Flash continued weak due to oversupply situation in the market. These factors led to a decline in our revenues and bookings during the quarter. However, this was partially offset by an increase in NAND Flash bookings, which outpaced those for NOR for the first time. Last quarter, we announced our flagship NAND product, NF150S (ph), to further penetrate the NAND market. This quarter, we recognized the first revenue from this product and customer acceptance of this product is growing. Our wafer-level burn-in product will experience continued growth during the first quarter of 2005. Today, we completely compete primarily with manual wheel-based (ph) probe card applications, which do not support large probe areas or a high temperature required by our customers. A year ago, we announced this product line to the marketplace and we have begun to see a broader acceptance across the delivery industry as more and more customers are unloading (ph) portions of stress and reliability tests towards the wafer. This revised test wall is currently exercised mentally (ph) for KGD, known good dye applications, but the products go over (ph) to commodity and DRAM devices as well. Additionally, the sharp rise in packaging costs for DDR2 devices has significantly increased the yield benefits and value proposition for this product. And customers are requesting higher parallelism, dual touchdowns and a wider range of temperatures for the applications, all of which have been the foundation of FormFactor's success in memory wafer tests.

We saw a noticeable pickup in our Logic bookings this quarter. As we have said in prior calls, this was due to the additional capacity from the ramping of our new facility. As part of our strategy to broaden our product line beyond DRAM, and to lower the cost of test for our customer, once again, the Logic market remains a key area of focus for FormFactor.

Our operating performance exclusive of new factory startup costs exceeded our expectations, as gross margins improved in the first quarter. Our programs to reduce lead times and internal cycle times are beginning to pay dividends as customer order activity has picked up. We are in the final stages of preparing our new factory for production. Production in our old site continues on a 7 by 24 basis and still accounts for the majority of our current revenue. One of our top priorities is to ensure the programs we're putting in place in our new factory will facilitate our competitiveness and enable us to be the lowest-cost provider in the industry.

I would like to thank our employees for their hard work and dedication in making this a successful quarter. I would like to turn the call over to Ron Foster, our CFO, who will provide the first-quarter results and our guidance for the second quarter of 2005.

Thanks, Joe. Revenues for the first quarter were 51 million, an increase of 4.9 million or 10.5% over the fourth quarter of 2004 and an increase of 13.9 million or 37% compared to the same period of last year. From a geographic perspective, we saw strength across most regions, especially Korea. As Joe mentioned, DRAM revenues grew 31% from the fourth quarter fueled by the transition to DDR2. Flash revenue fell 49% compared to the fourth quarter; a softness in the NOR market continued. Wafer-level burn-in continued to grow, an increase of 130% over the fourth quarter. A breakdown of revenues by major market and geography is available on our Web site.

First-quarter bookings grew 24.9% over Q4 and 1.6% over Q1 2004 to $50.9 million. Geographically, demand was strongest in Taiwan and Korea, with all other regions showing strength except Europe.

From a market perspective, bookings increased in every area except Flash. DRAM grew 20% over the fourth quarter to 38.5 million, due in part to the reacceleration of DDR1 bookings driven by 90 nanometer conversions. Additionally, Logic bookings grew 269% from last quarter to $6.7 million. Flash bookings declined 10% from the fourth quarter, despite a pickup in the NAND bookings.

Our turns ratio, based on customer demand, continued to increase during the quarter from 46% in Q4 to 53% in Q1. As we projected last quarter, our turns ratio normalized during the quarter and our backlog has returned to our standard six to eight weeks.

Going forward, we expect our turns to stabilize at about 50 to 55%. While normalization of our backlog and higher turns ratio demonstrate that we're making progress towards our goals of reducing leadtimes and improving on-time delivery, backlog will become less of an indicator of future business as customer responsiveness improves.

GAAP gross margins for the first quarter were 43.4%, up from 42.2% in the fourth quarter and down from 51% in the first quarter of 2004. Increased revenue (technical difficulty) improved efficiencies, partially offset by increased nonrecurring new site costs, accounted for most of the sequential improvement in gross margin, while new factory startup costs resulted in the year-over-year decline.

First-quarter startup costs increased by 1.5 million over the fourth quarter, to 4.4 million, slightly higher than our guidance of 4.2 million. Gross margins, net of new factory startup costs, was 52%, up from 48.6% in Q4. Startup costs per share matched our guidance of $0.07 per share.

As a reminder, we have republished the analysis on our website that models the financial impact on 2005 of the transition to our new factory. This schedule is consistent with our communicated timeline and phaseout of production at our old factory in the third quarter of 2005. Our troughs (ph) continue to trend in line with the assumptions laid out in the model for the second and third quarters. As indicated in the analysis, the model assumptions presented are estimates based on 80% capacity utilization and do not constitute guidance.

Operating expenses for the quarter were $15.5 million, up from 14.2 million in the fourth quarter of 2004. The majority of the increase was related to higher personnel expenses and distributor commissions related to higher shipments to Asia.

Legal and compliance costs were flat sequentially at 1.9 million as we continued with our ongoing litigation and Sarbanes-Oxley compliance work. Operating expenses excluding stock-based compensation as a percent of revenues were 29.2% in the quarter, down from 29.9% in the fourth quarter of 2004, as R&D and SG&A grew more slowly than revenues.

Operating income for the first quarter was $6.7 million, an increase of 1.4 million, or 28% over the fourth quarter of 2004. Operating income for the first quarter included $754,000 for the amortization of deferred stock-based compensation. Interest income for the quarter was 816,000, reflecting an average yield at 1.5% for primarily tax-exempt investments. Other income for the quarter was 87,000.

The tax rate in the first quarter was 35%, up from 19% in the fourth quarter and down from 39% in the first quarter of 2004. The increase in Q1 versus Q4 tax rate was related to a favorable rate adjustment in the fourth quarter of 2004, which adjusted down the tax expense for the 2004 fiscal year.

Net income for the quarter was 4.9 million or $0.12 per fully diluted share, versus 5.8 million or $0.14 per fully diluted share during the fourth quarter of 2004 and 5.1 million or $0.13 per fully diluted share during the same period of 2004.

Earnings per share in the fourth quarter of 2004 and the first quarter of 2005 included a decrease of $0.05 and $0.07 per share, respectively, for new factory startup costs.

Cash in marketable securities were 191.7 million in the first quarter, down 2 million from the last quarter. DSO was 34 days during the first quarter of 2005, versus 50 days in the fourth quarter of 2004 as a result of more linear shipments and a larger portion of our sales going to customers with shorter payment terms. Net inventories decreased by 878,000 to 10.4 million during the quarter.

Cash flow from operations during the first quarter of 2005 decreased 1.9 million compared to a $19 million increase in the fourth quarter of 2004 and a $739,000 decrease during the same quarter of 2004. The sequential decrease in the first quarter was due to the payout of fiscal 2004 bonuses, a reduction in the tax benefit from option exercises and an overall increase in spending to support revenue growth and our new factory.

During the first quarter, we spent 2.6 million in capital expenditures, down from 14.3 million in the fourth quarter. 3.1 million was received from the issuance of stock associated with the exercise of employee stock options, down from 5.9 million in the fourth quarter.

Headcount increased from 599 in Q4 to 616 in Q1.

With this, let me give you our guidance for the second quarter of 2005. We expect revenues for the first quarter to be 51 million to 53 million. GAAP operating income is expected to be approximately 11 to 12%, including 9 to 10% of new factory startup costs. We expect stock-based compensation of $800,000 for the second quarter. We target GAAP earnings per share of $0.11 to $0.13 per fully diluted share. We expect new factory startup costs of 4.9 million, which converts to $0.08 per share. Now, let us open the call for questions. Operator?

Thank you. (Operator Instructions). Jim Covello, Goldman Sachs.

A couple of quick questions. The higher revenues and the higher bookings, was that all coming from DRAM, or was there better-than-expected business levels from Flash or Logic during the quarter as well?

Jim, it was mostly DRAM, but also, a little bit in the other areas. We said that Flash was down.

Yes, I just wasn't sure if it was down less than you might have expected.

It was about as we expected.

Now, relative on the margin side, can you help us understand the difference between -- whether it be, I guess, gross margins -- in the DRAM business versus the Flash business and then the difference between NAND and NOR?

Jim, this is Ron. We don't specifically break down the gross margin structures between our various businesses. They all flow through one factory process. So we view that as one aggregated pool in the way we communicate it.

Then, I guess you could ask it the same way or a different way by asking on pricing on the Flash versus the DRAM products. I guess the question I'm trying to ask is, as the mix shift changes a little bit to NAND, does that have any kind of impact on margins, one way or the other? That's what I'm getting at.

Jim, this is Jens. So, as you look at it, right, the pricing is driven by the total cost of normal benefits in the different product lines. So what you're seeing, obviously, in DRAM, we're driving very high-parallelism tests of very fine features sizes. You heard us talk about the NF150 product on the Flash side, which brought an equal value proposition in the Flash market segment. So, at this point in time, as we said and repeatedly said, I think, on prior calls, is there is not a huge margin divergence here between the product lines, and it's very much driven by the individual value proposition on verification.

Bill Lew, Piper Jaffray.

Just a quick question on your revenue guidance for the second quarter. It sounds like DMS is progressing pretty well. You've got a big booking, or a big dollar amount in bookings in Logic. So why are revenues flat quarter-over-quarter? Is it mostly because Flash is weak right now -- NOR Flash?

This is Ron. In terms of the guidance of 51 to 53, our guidance range is up one -- a couple of million. So it could be flat to up in terms of the guidance. We see continuing strength in DRAM, and if you look at our book-to-bill ratio, it's about 1.02. So that's showing signs that we're in the same range to slightly up. That was the reason for our guidance.

Edward White, Lehman Brothers.

I was wondering if you could talk a little bit more about the Logic strength -- the kind of applications -- whether there is any parallel testing or whether it's mainly flip-chip logic, and how you would expect that to play out as you look at the next few quarters?

Okay, Ed, I'm going to take a shot at that. So, you may recall on our last earnings call, we were talking about the progression of the Logic business and that our focus was initially, as we're bringing the new factory up and new capacity, that we would be ramping back up existing customers. So that's what you have been seeing here, was an increased intake of orders during the first quarter. We've seen a strong resumption of our business for both microprocessors and chipsets. And on top of that, we continue to work on new customer engagements. We had first (ph) shipments into Logic customers outside of microprocessors and chipsets.

Peter Wright, CIBC World Markets.

Congratulations, guys. My first question is, focus on the DRAM market. How are you guys bundling DRAM industry CapEx through calendar '05? Do you see your customers loading spending into the first half, or are you expecting somewhat linear spending, specifically in light of some of the upward revisions coming out of the Taiwanese DRAM guys? And secondly, in light of the litigation with Phicom going pretty well, your way, are you seeing possibly one very large customer shifting some incremental business back your way?

On the capital spending, a lot of the front-end equipment has been delivered to some of these fabs. But our situation is really dependent on the pace of the transition and the startup and capacity ramp of these customers. And so that's occurring in more of a linear fashion. With DDR -- mentioned DDR1 accelerated a bit as we go to 90 nanometer. 90 nanometer is going to be a strong driver of the second half. DDR2 is still peaking with respect to bid growth, and a lot of the 300 millimeter capacity that is currently being put in place won't come on until the end of the year. So it's much more of a linear growth situation in terms of the DRAM demand.

And the other issue is also -- we talk a little bit about, with respect to the mobile RAM, is shifting production, being able to go from a commodity DRAM to a mobile RAM production and moving your capacity around. That alone (ph) feeds our business, because you have to test these devices.

And let me -- first I would like to add to what Joe said, and normally, we get very, very busy to supply probe cards for capacity that customers can't put it (ph) in place, right? So that very much drives our business. Not necessarily waiting for capacity to be installed, right? We are filling the fabs that customers build and put front-end equipment into, right? And second, equipment. And of course, when customers put extra -- normally, when customers put more equipment in, right, then again we'll follow with our shipment (ph). So we are a little bit different from a capital equipment company.

On the litigation side, we make it our point not to get terrified and then react, but rather it's a really simple equation. We invest in R&D, we do evolutionary breakthrough technology and product development, and products of those efforts belong to us. Therefore, when we see that potentially those efforts are misappropriated by somebody, we go and defend our IP before it becomes a very large problem. So the short answer to the question is that we do not do IP litigation as a reaction to some big loss of business, right? We do it because we protect what's ours and what belongs to our shareholders.

Marawi Aberi (ph), J.P. Morgan.

So perhaps can you just tell me how your Flash revenues break down right now between NOR and NAND Flash? And also in previous calls, you've talked about market share gains in NAND. Could you let us know what your market share is there at this point?

So maybe I'm going to take a first shot at this. So, as we said in our opening remarks that NAND and (indiscernible) outstripped the NOR business this quarter. And as you recall, historically, we started our penetration in the NAND market on the NOR side; that has been historically the higher percentage of our revenues. I think everybody is informed about the weakness that the NOR market, that it itself carries. On the NAND side, I think we have made continued progress. I think on the last call we talked to you about a few trigger points with respect to growth resumption on NAND Flash. Obviously, bid growth on NAND Flash is very strong. And we talked about our four touchdown, 300 millimeter product and F150, which also is a two touchdown, 200 millimeter solution. There's still a lot of the NAND capacities at 200 millimeter. So we had first revenue out of this product. We see quite strong acceptance of this product. And that product is going to be the backbone of both revenue growth and market share gains on the NAND side.

The only other thing I would add to that is we expect the Flash business to get better for us with this F150 product, because bid growth is strong and the market is moving more to our capability of higher levels of parallelism. So we think we've got the sweet spot. We should see a better situation going forward.

Medhi Hosseini, Friedman Billings Ramsey & Co.

I have two questions. First, I want to get a better understanding of how DDR2 ramp is coming. There's a significant capacity that is coming online; sounds to be just about 30% of the mix is going to be DDR2. Are you suggesting that they already purchased -- or have the first shipment of DDR2 products? Are they going to wait until additional capacity comes online, or are we just hitting an air pocket? And the second question has to do with tax rate. If you could clarify if the tax rate is going to go back up to 36%? Thank you.

Okay, Medhi. I want to answer the DDR -- DDR2 question and then Ron is probably going to address the tax rate. So there is no air pocket in DDR2. As we've seen, this conversion is happening in full strength, and I think you heard on the call that the DDR2 bookings declined very slightly, while DDR1 had some more strength. You really see here I think there are some mix adjustments here. Also, there is some 90 nanometer shrink already happening ahead of the curve for DDR1 devices. So, please, I think what we want you to take out of this call is that DDR2 is happening very strong, but also 90 nanometer is already emerging, probably ahead of what we would have originally thought on the curve because everybody giving the worldwide pricing environment of DRAM is getting ready to reduce costs. And as you know, this time, more challenging situation is very favorable for our business situation. So we are seeing a lot of proving cycle right now in DDR2, 90 nanometer following here and then the mixtures as people really try to optimize their revenue per wafer and I think we used mobile RAM as an example. Right now, mobile RAM are priced by almost 2X over commodity DDR1 designs.

Medhi, this is Ron. To your second question, in terms of the tax rate, we had a 35% tax rate in the first quarter, and that is the projection for the fiscal year 2005 at this point.

I did want to add to the previous question, the first part of your question, and that is, you know, the course (ph) DDR to RAM basically is strong and it's on tryout. And as you know, FormFactor -- the main areas of focus for the Company is to continue ramping capacity in your old facility, but now we're entering the stage where we need to execute and bring in line and transition to the new facility. And, with capacity situation being where it is, we dedicate a big part of this precious capacity to customers who completely rely on us. And DDR2 applications, or some of the DRAM applications, when customers need to ramp capacity or they need to transition to new technology, we are often mission-critical, and that does in fact -- to address a previous question -- our ability to aggressively go after businesses where customers potentially could get by.

Mark Bachman, Pacific Crest Securities.

Judging by your comments and also by your published results, it looks like the Korea was quite strong this quarter. When I kind of look at your 10% customer list, though, I don't see one there from Korea. Can we assume that there is going to be a change in your 10% customers when you publish your Q?

This is Ron, Mark. Yes, there is one new 10% customer out of Korea.

Okay, and then, I'm sorry, on the NF150, can you describe that product a little bit more in detail? I wasn't sure if you were talking 200 millimeter, 300 millimeter, number of touchdowns, kind of that that's under -- that that -- count, things like that?

So I'll put some color around it. So, as you recall, we released a platform of TH150 (ph) to DRAM last year, and so that's a 150 millimeter square array size that allows physical four touchdown off a 300 millimeter wafer. So we targeted initially a DRAM application; from around the beginning of the year we ported this platform over to NF150, which is the Flash equivalent. So now what we can do with this product market, you can either do -- supply a four touchdown, 300 millimeter solution, or you can also, with a smart layout configuration, do two touchdowns with the same active area on a 200 millimeter wafer. And we're selling both applications.

Kevin Vassily, Susquehanna Financial.

Question for you, Igor. Maybe you could provide a little commentary around the Harmony architecture? What are the major improvements that it offers over the current platform? Is there anything about the current alpha customer you are working with on this that made this platform particularly more attractive than your prior platform?

For the first part of your question, as you can guess by now, we're pretty excited about this. And this new architecture does apply to all of our product lines across the board -- support platforms. And the beauty of it is that it allows us to start scaling in the (inaudible) active area and eventually as we said, 150,000 (ph) and 300 millimeter, and expanding it and doing a 500 millimeter faster wafers. We'll probably be able to furtherize (ph) that in the future. With architecture that is fundamentally scalable, the beauty is that it also allows us to use all our other technology. So we don't need to sit down and redo everything in the Company. So you can focus on the short-term developments, you can focus on new technologies that allows you finer pitch testing, and those will be incorporated in this architecture and get expanded in exit (ph) area that we can test. So it applies to answer the second side of your question, this is first evaluating customer, but it is applicable to low (ph) customers and it's applicable to NAND, Flash, it's applicable to NOR Flash, it's applicable to DRAM, it's applicable to wafer-level burn-in.

Tim Schultz Milander (ph), Morgan Stanley.

Congratulations on the quarter. Three quick questions, if I may, the first to relate to bookings. Clearly, project bookings were very strong. Can you just give us a sense, as you look forward, do you expect that to be an extended cycle in your Logic activities? There is an extended retooling cycle, or is that more likely to be lumpy?

I think I'm going to somewhat repeat what I answered an earlier question here, Tim. Right now, I think what we're doing is we're bringing up new capacity; we're ramping back up existing customers. You may recall during last year, we somewhat deemphasized some of the Logic business for capacity reasons and so, we're right now consistent with the strategy of reengaging with those customers and ramping with existing customers. So, here, I would not necessarily talk about the tooling cycle, because as you know, our overall Logic strategy really deals with gaining market share and new customer engagements, and as I mentioned earlier, that we have shipments to customers this quarter already outside of microprocessors and chipsets.

Patrick Ho, Legg Mason.

One clarification question and then a basic fundamental question. First, on the Logic product margins versus the memory product margins, at a high level, can you just kind of give what the difference is and how much you can improve the -- I guess the Logic product margin over time as it starts ramping up? And the clarification is on SG&A. Was that bump-up due primarily to additional legal costs in the March quarter?

Patrick, this is Ron. With regard to your first question, again, we don't break down margins by individual product categories. But, in general, they are fairly typical structures and go through the same manufacturing process. With regard to SG&A, some of the growth was related to legal and compliance costs, I mentioned in my comments, and some was just to staffing increases.

Stephen Rosten (ph), Glen Capital Management.

A question on your 150 product for NAND Flash. What do you think it will take for your customers to embrace this in terms of changes in their business or changes in the environment, so that you can take a similar type of share there than you have in other markets?

Steve, I think there's really no infrastructure changes required on this product. This goes right into the existing test cells and testing platforms and that's why we've seen fairly fast acceptance and adoption of this product. So there's no additional infrastructure requirement or acceptance criteria in there.

Another way to look at it, again, it plays into more parallelism with higher bid growth means we do seem to call up (ph) a test for the customers. So it plays right into our hand, in terms of this product offering. And now, we're going to have more capacity to be able to address this particular segment.

Tim Schultz Milander, Morgan Stanley.

Just wanted to follow up on the DRAM site. You had referenced that MRAM had been very strong, particularly in Japan, and then you sort of gave some granularity on DDR2 and DDR1. If you strip out the MRAM, how does the DDR1-DDR2 picture look? Because you said DDR2 is down slightly and DDR 1 was up. Do the two sort of offset each other, or would they both have been up in aggregate?

No, I mean I think overall you saw the continued strength in the DRAM side as the architecture is growing. So the mobile RAM -- I want to make sure that it is not confused with MRAM or other architectures, right? Which is magnetic RAM. So mobile RAM for portable digital consumer applications is strong, and as I mentioned earlier, I think the strength is driven by two things. Number one, as we repeatedly reported in earlier calls, we've seen growth in this area of digital consumer applications consuming more DRAM devices. And the other one is, I think, we're in this seasonal pattern for the consumer side to strengthen and there certainly is a loss (ph) -- a better pricing premium on these devices at this point in time, due to the NAND.

And on the DDR1, DDR2, I think what is happening is that as we're approaching DDR2 bid parity and we're saying that it will be probably at around Q3, you'll see customers pushing very, very hard on 90 nanometers. And when customers are preparing for that, they normally do it on the previous architecture, so they're using -- it looks to us like DDR1 to get ready for big-time DDR2, 90 nanometer push.

Bill Lew, Piper Jaffray.

Just a couple of follow-ups. It seems like wafer-level burn-in is doing fairly well here. I'm just wondering if you could help me with going forward, what can I look for as a catalyst or maybe a technology change that would really drive more wide-based adoption. You have the hockey stick. What's out there that that could drive that and what can I be looking for?

I think we're very encouraged about what we've seen through the quarter on wafer-level burn-in, Bill. And I think there's multiple aspects that drive that. You may recall at the beginning we talked very much about the known good dye flow, KGD, right, and (indiscernible) modules and very much driven by mobile applications. So that was the first lack in adoption here. Now, what we're seeing on top of just the KGD flow is that also normal commodity DRAM devices are adopting increasingly wafer-level burn-in. So you have management reasons. Because if you look at right now the DDR2 device, and going forward the architecture is going to drive it even further. The packaging costs in the back end are growing rapidly. So a DDR2 package is pretty much 2X over that of a DDR1 standard T-sell (ph) package. And this is driving more and more tests from across the wafer level as people try to go through a liability and labor redundancy repair at the wafer level to avoid a packaging bad buy. So I think this is the next catalyst for this, and then obviously, this is driven out of the commodity side of the DRAM market, which is sizably larger still, and will for a long time, be larger than the portable application.

Peter Wright, CIBC World Markets.

One of the transitions we've seen is all of the Flash guys move their test capacity or fill their Flash capacity with older-generation DRAM testers. And I'm wondering if that's an opportunity for you guys that you see in the Flash market that seems to be played to be able to create a revenue opportunity for you if there is a new probe card that needs to sit on that tester?

I think that what you're seeing actually really is that people try to have maximum flexibility between the two product lines. Obviously, where older testers are deployed, we have been historically able to expend resources and performance of the tester with some of our proprietary technology. And, where such opportunity arises, we will certainly take advantage of it.

Yes, this is Igor to add to that. You are right. What happens is that normally customers, at least some customers, what they do is, depending on ASPs (ph), that drive for a particular device, that drives cash total revenue per wafer. They'll shift capacity, their manufacturing capacity, from Flash to DRAM depending on what, as I said, it looks like. And in order to do that shift quickly, customers typically will have to maintain both suites of probe cards. And so you are right, in a way, that we're doing more probe cards per tester head in those situations.

Marawi Aberi, J.P. Morgan.

Can you confirm at this point your revenue guidance represents actual demand and that you guys are not constrained by capacity anymore in terms of your revenue?

Marawi, this is Ron. I will take a first shot at that question. We have been capacity-constrained for some time and are still operating out of the old factory and only beginning to transition into the new one. So we are still in a situation where, until we get production up in our new facility, we will not be able to deal with the volumes of demand that we would otherwise be able to accommodate.

I would add that this quarter, we will be at 20 to 25% of production in our new facility. So we still have work to do to make that happen. But I think this is very consistent with the analysis that we've provided. It puts the capacity in the range of our revenue guidance. So we feel pretty comfortable. I think as the factory comes on stream, we will have more and more capacity.

Medhi Hosseini, Friedman, Billings Ramsey & Co.

Going back to DDR topic. Are you suggesting that that's (ph) basically taking their underutilized 200 millimeter, 110 nanometer DDR capacity and converting that to 90 nanometer on the same wafer size?

No, I don't think that's -- that's an accurate statement, Medhi. I think what you see right now is we still have chipsets in the market that support both DDR1 and DDR2, right? So if you look at prior conversions, we've seen these swings at the end of the transition between new and replacing architecture. And what -- fundamentally what people are doing is they are driving to 90 nanometer to bring their costs down, and they are doing it with existing devices, as Igor mentioned. It's a safer bet to ramp a new technology with an existing design, because you eliminate unknowns, and you understand your yield to earning better. So I think we see some of that going on. As you know, there are chipsets now coming into the marketplace that are solely DDR2, right? So this means as we go forward, especially in the desktop environment, we will announce the new chipset platforms are all solely DDR2 in this setting.

So, let me just say, Medhi, so there's no confusion. What happens is around this new architecture now, you are ready, really, to basically fundamentally start transitioning. Two ways -- you have some new testing methodologies that you are now comfortable with. You've brought some different testing equipment and all these other things. And now, it's all in place, and the wafer used to increase capacity transitions to 90 nanometers and gets a lot more bits out of every wafer without having to go and buy additional full capacity. So that's what normally happens to it towards the big parity with any architecture.

Okay. That completes our first-quarter conference call. I want to thank you all for joining us. Once again, you can hear the replay of this call on the investor portion of the FormFactor website, and we're looking forward to seeing you at some of the upcoming conferences and on our next conference call. Thank you.

Ladies and gentlemen, this concludes FormFactor's first-quarter 2005 earnings conference call. You may now disconnect.