FormFactor Inc (FORM) 2007 Q3 法說會逐字稿

Good afternoon ladies and gentlemen and welcome to the FormFactor's fiscal 2007 third quarter Conference Call. My name is Jen, and I will be your coordinator for today.

(OPERATOR INSTRUCTIONS).

Now I would like to introduce Ms. Annie Leschin of FormFactor Investor Relations. Ms. Leschin, you may begin.

Good afternoon and thank you for joining FormFactor's third quarter 2007 Earnings Conference Call. With me on today's call are Igor Khandros, Chief Executive Officer and Ron Foster, Chief Financial Officer.

Igor will provide a summary of our third quarter performance, review market segments and provide an update on outlook and long-term strategy. Ron will then take us through the preliminary financials, operational details and provide guidance.

I would like to take a moment to mention that during the fourth quarter of 2007, the Company will be presenting at the Lehman Brothers Technology Conference in San Francisco on December 5. As other events and details become available, we will make additional announcements.

Finally, before I hand the call over to Igor, I will review our Safe Harbor Statement.

During the course of this conference call, we will make forward-looking statements within the meaning of the Federal Securities Laws, including statements regarding FormFactor's growth and financial performance as well as our strategic and operational plans. These 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 timing and results of completion of the Company's internal financial review, including the risk of changes in the reported amounts or guidance reported in our press release or on this call; the possibility of additional third quarter charges with respect to prior periods and the possibility of restatements of prior periods' financial statements; the Company's ability to effectively drive its product development plans and introduce and qualify new products that meet customers' testing requirements in Flash, DRAM and logic; the Company's ability to innovate and develop and deliver the next-generation of testing technology; the demand for certain semiconductors devices; the rate at which customers adopt the Company's newly released architectures, technologies and products; implement manufacturing capability changes; make the transitions to smaller nanometer technology nodes; and implement tooling cycles; the Company's ability to continue to ramp design and manufacturing capacity and increase efficiencies in its operations; and the Company's ability to obtain other cost advantages from its expansion of manufacturing operations in Singapore.

This Company assumes no obligation to update the information in the press release, to revise any forward-looking statements, or to update reasons actual results could differ materially from those anticipated in forward-looking statements.

Please refer to the Company's recent filings on Form 10K and 10Q for additional and more detailed information regarding the relevant risks and uncertainties. FormFactor undertakes no obligation to review or update any other forward-looking statements.

Finally, a breakdown of revenues by market and geography and a schedule reconciling our GAAP and certain non-GAAP financial guidance with respect to FAS 123R stock compensation expensing is available on our website.

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

Thank you, Annie. FormFactor delivered another record quarter with revenue reaching a historic high of $125.3 million for the quarter, a 10% sequential increase over the second quarter. Ron will discuss our preliminary financials and the reasons why we're only providing limited information today. We are very focused on getting the financial issues resolved and bringing our financials up-to-date.

The market for advanced probe cards continued strong in the third quarter, with DRAM once again leading the way. FormFactor's growth was driven by the record high DRAM, Flash and known good die businesses, as we continue to meet the mission-critical needs of our customers.

Design activity for new and existing products in DRAM, Flash, and logic also accelerated in the quarter as we worked to penetrate new markets and expand our presence in existing ones with markets transforming advanced probe cards.

We continue to expand our manufacturing operations as we position several new products into production phase. We've furthered our production into Singapore with our groundbreaking announcement this quarter, and continue to invest resources in design ad engineering.

The DDR market was robust in the third quarter, driven by the continued ramp of 70 nanometer technology node, and the transition to one gigabit DDR2. With anticipation of higher DRAM demand for Vista systems in the second half of 2007, DDR manufacturers have been aggressively adding capacity and shrinking to 70 nanometers. The transition to 70 nanometer probe cards has moved from the early stages to approximately one-quarter to one-third through this tooling cycle. By year-end, we expect to be roughly 40% through the 70 nanometer cycle, leaving the majority of opportunity still ahead.

We're also excited by the fact that we have already shipped probe cards for 65 nanometer designs, which is the beginning of the transition to the next technology node. These trends were reflected in our results this quarter, with robust 70 nanometer and 1 gigabit DDR2 demand. This was the first quarter that our 1 gigabit bookings exceeded those for 512 megabit.

We experienced higher than expected demand for our DRAM Harmony XP full wafer contactor product, which offers superior test cost benefit for 1 gigabit or higher density devices. Having shipped Harmony XP to 3 customers, initial results demonstrated superior industry-leading wafer-testing capability. We believe that our Harmony XP product is the best positioned and most advanced probe card in the market, able to address the tightest test [base], smallest test size, and highest probe density in (inaudible) greater than 40,000 probes today.

We believe this product will accelerate the future move to one touchdown in DRAM wafer testing, while the growing complexity of DRAM will eventually limit this capability of competing probe card technologies. We have already received additional production orders for DRAM Harmony.

As we continue to make strides in the qualification of our Harmony product, the volume ramp is not keeping pace with the combined strong demand for DRAM and NAND Harmony. In order to meet this demand, we made a decision during the third quarter to add more production capacity for assembly and testing of Harmony cards, including clean room space, equipment and personnel.

Our efforts in the near term are focused on successfully transitioning the Harmony platforms from the low volume engineering-assisted production phase into high-volume manufacturing phase.

Finally, design activity around DDR3 continued to gain momentum this quarter. We saw a significant increase of DDR3 bookings to approximately 5 million, primarily for 70 and 65 nanometer 1 gigabit DDR3 designs as manufacturers prepare for production ramp in the second half of 2008. DDR3 will drive an approximate 20% increase in the number of pins per die over DDR2.

We believe this will potentially inflate to as many 100,000 probes per one touchdown wafer contact. This will again play to FormFactor's strength. We anticipate a similar coexistence of architectures that will witness in this transition from DDR to DDR2 in 2004 and 2005, which will benefit FormFactor and (inaudible) probe card market. DDR3 is shaping up to be a major growth driver for the second half of 2008.

Our known good die business experienced its best quarter yet. With device speeds rising and new high-speed testers entering the market, customers will increasingly higher speed probe cards. This quarter, we saw higher adoption of our high-frequency test at probe product was largely driven by the demand for (inaudible) mobile devices, such as NOR, specialty NAND, mobile RAM, and PSRAM.

Additionally, customers are beginning to see the benefit of performing more tests on the wafer for individually-packaged die, such as commodity and graphics DRAM. In such cases, testing speeds are approaching 500 megahertz for early detection of speed-related defects. As a result, we are seeing increased design activity for our high-frequency HFTAP product. We expect 2007 high-frequency probe revenue to exceed 2006 by more than a factor of two.

The wafer level burn-in at a slower quarter due to the timing of customer orders, we expect continued strength for the remainder of the year with the ramp of 70 nanometers. Additionally, we are seeing the market transition to full wafer contactors for wafer level burn-in applications. As I recently announced the shipment of our new Harmony wafer level burn-in product to multiple customers.

With record scalability and high pin count of 40,000 probes, this is the first industry for one touchdown 300mm probe cards for wafer level burn-in. This card can potentially provide an order of magnitude improvement in (inaudible) for some manufacturers, resulting in compelling test core savings. We are confident that known good die will continue to be an important contributor to FormFactor's 2008 and long-term growth.

Our Flash business had a particularly strong quarter as NOR reached record levels due to customers' production ramp of 90 nanometer and 65 nanometer products. Our continued success in NOR is a testimony not only to our technological leadership, but also to our ability to deliver consistent test cost reduction to our customers year-over-year.

Progress in our NAND Harmony OneTouch products continued during the quarter, as we received notable orders for our Harmony OneTouch 300mm cards from 2 customers. Initial performance of this card, including high volume manufacturing upside and measured yield improvement during the device ramp stage have been very good, but we are working to win more designs at existing customers, and engage new customers. The manufacturing ramp on Harmony products are still behind current customer demand.

The third quarter investment in additional Harmony manufacturing capacity mentioned only in reference to DRAM will also assist our Harmony and RAM due to the commonality of the Harmony platform. The Company is very focused on aggressively expanding Harmony capacity and decreasing lead times.

FormFactor's logic SOC business declined this quarter due to the tooling cycle of our primary flip-chip customer. Flip-chip is a growing market. We continued our efforts to expand our customers' base this quarter, engaging in shipping costs to new customers in graphics and telecommunication segments.

In the wire-bond segment, we continue to see the migration to advanced probe costs requiring finer [test page] capability. In the next few years, more than 90% of the ICs shipped will stay wire-bonded, providing a large market opportunity for FormFactor.

Last quarter, we introduced our high temperature, high parallelism wire-bond product for automotive application. This quarter, we received production orders for this product from key customers, which resulted in the doubling of our wire-bond revenue. We continue to penetrate this market with our newly-announced revolutionary true scale product and acquired two wire-bond customers in the mobile communications and digital consumer segments during the quarter.

True scale will address the limitations of conventional SOC probe technologies that cannot scale to below 50 micro and [test page] of high parallelism. True scale is designed to increased manufacturers' throughput and lower [test scores] while supporting technology roadmaps, with [test] pages expected to reach as low as 30 micro within 1 to 2 years. We expect to make more announcements related to our wire-bond logic business throughout next year.

Key indicators for advanced probe card market remain positive. We believe that the overall market for advanced probe cards will grow in the mid-20% range in 2007, in line with our projection early in the year. Year-end will once again be the primary driver of the market, given the increase in big growth estimates, with additional contributions from Flash and logic due to the adoption of advanced technology.

FormFactor remains the market leader given our strength in DRAM as well as NOR, and our emerging presence in the NAND and logic markets. NOR and (inaudible) will become a key contributor to FormFactor's growth. Our continued evolutionary and revolutionary new product efforts are designed to assure FormFactor's long-term leadership in the advanced probe card market. We expect continuing growth in 2008 with year-over-year advanced probe card market and FormFactor's business outlook for 2008 on our next quarter earnings call.

In summary, the third quarter of 2007 was another very strong quarter for FormFactor. We are very pleased with the Company's achievement of record revenue, expanding factory capability, and continued new product development. I would like to thank the entire FormFactor team for their hard work and efforts in making this another successful quarter.

I will now turn the call over to our CFO, Ron Foster, who will elaborate on operating results and financial performance for the third quarter, and provide guidance for the fourth quarter.

Thank you, Igor. Let me begin by discussing why have not issued our full customary financial information today. As we were closing the books for the third quarter, we discovered an issue related to our inventory evaluation practices. It appears that in certain prior periods, in the current fiscal year and prior years, the approach followed for evaluating inventory and establishing valuation reserves was inconsistent with our accounting policies. While I cannot go into the details of the ongoing review at this time, I can provide some perspective.

In our business, where every product is custom for each customer's design, most of our inventory consists of custom parts. We have a comprehensive process for assessing the value of that inventory in light of projected customer demand. The failure to follow that process could mean that the previously-recorded inventory values would need to be adjusted up or down, with flow-through impacts on gross margin, operating margin, and net income.

We have been working intensely on getting our arms around the problem, including identifying the effective periods and quantifying the impact. We have made a great deal of progress in this effort. However, because our internal review is not yet complete, we decided that the prudent course of action was to make the more limited announcement that we announced today.

We are prepared today to discuss third quarter items that are not affected by inventory evaluation, such as revenue, bookings, accounts receivable, and cash. In light of the work we have done, we are comfortable in guiding third quarter operating margin and earnings per share, at least within the guided range provided in July without giving effect to any adjustment required to be made related to earlier periods.

We are not, however, at this point able to provide definitive numbers for items that are affected by inventory evaluation. So when will we be able to release full Q3 numbers? Our review of prior periods is continuing, and we are doing everything possible to complete the process as soon as practicable. The Audit Committee of our Board and our independent auditors are, of course, fully engaged in the process as well. Once we have determined the periods affected and the magnitude of the potential adjustments to inventory, gross margin, operating margin, and net income from the amount s previously reported, we will know whether we can simply do a one-time catch-up adjustment in our third quarter numbers, or whether a restatement of prior periods will be required.

Going forward, we will apply lessons learned in this review. We are taking a fresh look at our inventory evaluation process to ensure inventories are appropriately valued, and we will make the appropriate modifications to our procedures and compliance processes. We do not expect that this review of our processes will have any impact on our long-term financial model.

With that as background, let me turn to the third quarter items that we can discuss today, and then provide you with our guidance for the fourth quarter.

Revenues for the third quarter increased for the eleventh consecutive time, reaching record levels at $125.3 million, up 10% over the second quarter of 2007 and 29% over the third quarter of 2006. DRAM revenues increased 3% sequentially to $82.2 million and 17% over the same period last year. DRAM revenues accounted for 66% of total revenues in the quarter.

Flash revenues grew 60% from the second quarter to $32.2 million and 93% over the third quarter of 2006, representing 26% of total revenues in the third quarter.

Our strong performance in Flash this quarter was primarily driven by strength in NOR Flash sales due to an ongoing production buildup from a key customer. Additionally, we saw strong HFTAP growth from KGD applications. While we did see some traction in our NAND Harmony OneTouch product in the quarter, overall NAND revenue declined due to the production restraints, as Igo highlighted.

Logic SOC revenue was $10.9 million in the third quarter, comprising 9% of revenues. This represented a decline of 21% over the second quarter, and an increase of 12% over last year. The sequential decrease was attributed to a decline from a key flip-chip customer.

Revenues from known good die products consisting of wafer level burn-in and HFTAP were strong at $22 million, an increase of 24% over the second quarter, driven by mobile RAM and PS ram. Year-over-year growth was 64%. The breakdown of revenues by market and geography is available on our website.

Bookings for the third quarter fell slightly to $124.6 million.

Although we are not providing gross margin, operating margin, or EPS figures for the third quarter, based on the work done to date, we are comfortable re-affirming our prior third quarter guidance that we will be at least in the 20.5% to 21% GAAP operating range, and the $0.38 to $0.40 GAAP EPS range. This range does not include the effect of any adjustments that may be required related to prior periods.

Given our limited preliminary financials, I will not be able to highlight expense metrics; however, I would like to offer you some color on a few other financial metrics. Cash and marketable securities increased $11 million over the second quarter to $537 million. Cash flow generated from operations was $2 million, as tax and bonus payments were made in the quarter, along with a $7 million, 30-year land lease prepayment for our Singapore site, and we spent $14 million in CapEx in the third quarter.

Headcount grew to 1,154 from 1,108 last quarter. The headcount increase was mostly driven by new engineer hires in R&D in both California and in Singapore.

DSO increased to 48 days in the third quarter, compared to 41 days in the second quarter, due to an increased mix of customers with longer standard payment terms.

The accounts receivable balance increased $7.6 million from the second quarter. We continue to run with a high current AR balance, consistent with prior quarters in the high-80% range. As Igor commented, demand for full wafer contactors to support the ramp of our Harmony DRAM and Flash products is exceeding our production capacity. We plan to double Harmony capacity in the first quarter of 2008. We expect the continuing Harmony production ramp to meet demand in the second quarter of 2008.

We had the groundbreaking ceremony of our Singapore facility on September 21, 2007; we are actively hiring personnel to support our manufacturing efforts in this region, and have put in place a highly-experienced Executive Team. As previously communicated, we expect to fund our expansion into Singapore through cash flow from operations, and within the range of our annual target model.

We are revising our plan, and now intend to ramp sales and customer support operations in the first half of 2008, and to begin production in early 2009. This, combined with the R&D cost-sharing payments we initiated in the third quarter will position to Company to benefit sooner from lower tax rats in Singapore associated with our intellectual property assets. As part of the migration plan, Singapore will likely make a one-time partial prepayment for US-generated intellectual property assets at the end of this year. The prepayment amount has not yet been determined.

Excluding the prepayment, we still expect the tax rate for the fourth quarter of 2007 to remain in the 39% range since the favorable tax-exempt status we have been granted in Singapore has made net loss carry-forwards unusable. However, with the early start-up of our sales operations in Singapore, we project our tax rate for 2008 to fall to the mid-30% range. As manufacturing volume ramps in Singapore, we expect that our tax rate will drop to the high-20% range in 2009, and fall below 20% by 2012.

Now I'd like to provide guidance for our forth quarter of 2007. As we have highlighted in the past, we do expect to see some fourth quarter decline in the consumer mobile segment, which is over 40% of our current business. We expect this to impact our performance in NOR, mobile RAM, and KGD. Though this has been offset in the past by specific customer tooling cycles, particularly the strong node transitions of penetration of a new market, we do expect relatively flat revenue performance this quarter considering the consumer market outlook and the Harmony production ramp constraints. The projected revenue range is $123 to $128 million.

As already mentioned, we have found it more difficult than planned to ramp Harmony to volume production, which is limiting our ability to meet increasing full wafer contactor demand. Consequently, we have put more capacity in place sooner than planned to support the volume ramp. This includes startup of a new clean room and associated manufacturing equipment and people in the fourth quarter. This will increase our near-term manufacturing costs until Harmony ramps to higher volume and comes down the startup cost curve.

In light of these factors, our income guidance for Q4 is in the range of 22% to 24% on a non-GAAP basis, and 17% to 19% on a GAAP basis, with the impact of stock compensation expense.

Our EPS guidance is in the range of $0.40 to $0.44 on a non-GAAP basis, and from $0.32 to $0.36 on a GAAP basis. This excludes the tax effect of any potential one-time prepayment for Singapore intellectual property that I previously described.

Now despite the Q4 margin decline, for the total year we do expect to achieve our target financial model performance of non-GAAP operating income of 25% once all the numbers are reported.

Now we'll open the call for questions; Operator?

Thank you sir. (OPERATOR INSTRUCTIONS).

Jim Covello, Goldman Sachs.

Good afternoon guys, thanks; couple of quick questions. First on the revenue, if I just plug in the mid-point of the revenue for the fourth quarter, let's say, that kind of gives you, that gets you growing about 25, 26% for the full year, which is roughly in line with the market. I think the previous indications had been that you guys felt comfortable outgrowing the market. Are you thinking about that Ron, or has there been a little bit of a change there?

Jim this is Ron; we commented on Igor's comments that we expect the market to be growing around the mid-25% range, around 25%. It might be a little bit lighter than that; it's hard for us to call, but it's in that kind of range. And you're accurate; we believe that we, with our guidance would grow 25% or a little bit better than that, but it's a closer spread than maybe we would have characterized a quarter ago.

And what do you think accounts for the closer spread?

Well, one thing we've talked about harmony production ramp constraints, which is limiting our ability to scale in the fourth quarter.

Okay; 1 or 2 more quick ones. DDR3, how many DRAM companies are starting the tooling cycle for DDR3? Is it 1 or 2, or is a little more broad-based than that?

Everybody is starting DDR3 design, and we've shaped probe cards for more than one design.

Okay, and then final question for me; on the accounting issue, what I'm trying to understand is when did this issue, how far back does this issue go in terms of it having been an issue the last couple of quarters, is this a continuation of previous policy, or it this a policy that was new to the last couple of quarters? I'm trying to understand how this slipped through the cracks for so long.

Sure, let me just cover a little more background on it. First of all, I want to acknowledge right up front we should have gotten this right quarter-by-quarter here, and while I can explain the problem, I'm not trying to excuse it. I also want to apologize in advance, but since the review is ongoing, I'm not able to specifically provide information until that review is complete.

But I can tell you that the review covers the 2007 period, and we are also looking at prior periods as well. And it specifically relates to the policy we have in place for valuation of inventory, and whether or not that has been properly applied period-by-period in the past, and that investigation I mentioned is ongoing.

And something that's important to understand is that our inventory business is quite a bit different from others. It's unusual. On the one hand, our risk is limited because the product for each customer is custom, and we therefore don't build products in advance of winning a design. But once our design has been selected by a customer, we actually regularly order components and build sub-systems in advance of receiving a purchase order, so we'll be in a position to meet demand, and those complements and sub-systems are by their very nature custom.

We then have to value this inventory each quarter in light of anticipated customer demand, factory yields, which also affect how much we buy, and other factors. It's actually a complex, multi-variable process, and in fact, better factory yields can end up generating unused and potentially un-needed material in our custom environment, even though it leads to lower product costs down the road.

So what I can tell you is that the problems (inaudible) relate to the valuation of the inventory. There's no issue with the units or count; it's the valuation of the reserve methodology related to the custom nature of our product. And we have not made any changes to our policy. We're validating that our policy has been applied properly in the past.

Does that help your question Jim?

I think so; thanks a lot.

Okay.

Mark Bachman, Pacific Crest Securities.

Yes, hi Ron; I just wanted to go back and follow up there to your answer. Are you saying this inventory is then is limited to your tenure, or does it date back to the previous CFO?

Mark, the review is still ongoing, and so what I can tell you right now is we're in an investigation and we're looking at 2007 and prior-year periods.

Okay, but can you be any more specific? Do you think it dates to the prior CFO?

Not at this point. We're still under investigation. I cannot answer that question specifically.

And is there a higher probability that this has been material or non-material?

We won't know until the review is over.

Okay, and your first indication that you saw of this problem, did you think that the adjustment was a negative effect on your margin structure, or was it a positive effect?

Mark, we're looking at each period; it can go either way because this is a complex valuation assessment. To give you an idea of maybe the order of magnitude and to help understand the comment I made to Jim earlier, if you look at our Q2 inventory, we had about $62 million of gross inventory, and we reserved nearly half of it. So we have a significant reserve level already in our constructs, but there's many steps you go through to make that valuation assessment. So we discovered some anomalies in our Q3 close process, and we're going back and checking not only 2007, but all prior periods to evaluate whether the policy was consistently and properly applied.

Okay, and then lastly, can you just talk about, are you being constrained by your suppliers, and then if you could give us an idea of where your, what you think your quarterly manufacturing run rate is? I ask because you've had a lot of bullish comments over the past several quarters, and especially so given the move -- in every segment of your business right -- and especially given the move into the 70 nanometer DRAM. But your guidance here has been considerably less than what The Street is looking for, and basically shows no growth quarter-over-quarter.

Right, so there are no 2 ways to say it; we are finding ramping Harmony more difficult than anticipated. And that's meaning of deciding in third quarter putting in additional manufacturing capacity for Harmony, and we're doing it earlier than we would have normally done considering the ramp of the business. So that's how, with this product, the risk number one is does it work. And now both the NAND and the DRAM we've delivered multiple probe cards to many customers, and when customers get these products and these products run in production, these are great products and people love them.

Now what the Company must do is to ramp it in production to the 4 and the 3 times, come down, and that we can make them in a cookie cutter fashion. So the big exercise right now is to get out of this, a lot of engineering involvement in production to basically just to our technician and production people a kind of control manufacturing.

So that's what we are going through, and that is, that will be the difference between what our outlook used to be and what it is right now. Having said that, we believe that we are past the main challenge and that is that product, the complex of this, does it work. And we have multiple, multiple evidence that it performs very, very well. It's just a matter of right now ramping it.

Okay, just one last one for me Igor; I mean if you go back to your last conference call, you were very adamant on that call that NAND wouldn't be a revenue driver here. And that's fine if you disregard that, and I remember Jim asking you a very specific question, if you needed NAND to get to your, to get to your stated guidance that you're in, and you were very clear that you didn't. So that means that in order to have flat guidance quarter-over-quarter, you were weaker across the rest of your segments there. And I guess that's what I'm driving at is where's the growth?

Well, we're guiding almost flat quarter-over-quarter, and the difference is the Harmony DRAM is behind what we then thought it would be today.

Okay thank you so much.

Thank you.

Tim Arcuri, Citigroup.

Hi guys, a couple of things; Igor even if I strip out the issue in December with the guidance, and I kind of look at what you're doing in DRAM, and I look at what the bit rough expectations were kind of entering this year for the DRAM business. We're kind of growing bit supply this year, 10% to 15% more than what I think folks though 9 or 12 months ago, and yet your DRAM revenues, even before these issues with Harmony, weren't really surprising to the upside. And if you take your DRAM revenues and you look at it on a per-bit basis, there's been a big collapse that I've heard about 4 quarters ago, and it's kind of continued for the last 4 quarters where your revenue really isn't growing as much with bit supply growth.

And I'm wondering, has something structural changed; is it a competitive issue; is it pricing; what do you think has happened?

You mentioned several things; one I'm not sure we have a problem with guidance. I'm not sure what that -- but getting, we are growing DRAM and we are growing significantly. That's the bottom line. We're going to 70 nanometer transition that we are saying that we're about one-quarter to one-third through. We're not changing the overall tooling cycle value to us, and if there were $600 million or so dollars. We are already seeing the beginning of 65 nanometer; DDR3 is going to be a very, very positive development next year, so we see DRAM continuing to be a strong revenue driver for this Company.

I am not sure what collapsed, what you also are referring to.

Yes I guess Igor what I'm -- let me ask it a different way. So if you strip out the December quarter guidance and just look at what's occurred 2 days through the end of September, and you look at the revenue of the DRAM that you're getting per bit, per DRAM bit, it's gone from, in '06 it was $0.70, $0.80 roughly, and since the middle of '06 it's gone down to, as of September it's about $0.45. And it was pretty consistent up until the middle of '06, and like I said, it's gone from $0.70 down to about $0.45.

So I guess just structurally what I'm wondering is are customers buying less of your cards and kind of getting more bang for their buck? Is that what's happening, or do you think there really is nothing's changed?

All right, I understand now. So if we're in business of reducing test costs. Test costs will be going down; bit growth will be going up; customers will continuously find a way to optimize their (inaudible) in a way that they'll be reducing times, and FormFactor will be continuously finding a way to help them to reduce test costs. So our revenue will not grow at the bit growth. It will never grow one-for-one.

Now in cases where multiple architectures co-exist, in case where design activity goes significantly up; for example, this year we only had one architecture, really DDR2; we still have a tail-end of DDR, but it's really one architecture. When you get into a situation when you have 2 architectures co-existing, you will get increase of acceleration to DRAM probe card revenue growth.

But please do not expect us to grow with the bit rate; that is not possible.

Of course, yes, so I guess just to kind of follow up on that and just close it, so what you would say is that in '06, in the last half of '06 and '07, you basically had DDR2 ramping whereas in prior periods where that number held consistent, you were ramping multiple architectures.

In '06, you have situations where DRAM suppliers were finding that DDR1 was selling at some point at a higher price than DDR2, and that drove people to order redundant set of probe cards, so they held them in hand for the likelihood that one or the other generates higher revenue per wafer. So that was, so we grew in '06 55%, right? That was one of the contributing factors to the growth.

I see.

As we enter '08 and you have multiple architectures co-existing, that again can be a positive factor for revenue. But it's not as simple an equation as just latching onto bit growth.

Right, okay Igor, thanks a lot.

Thank you Tim.

CJ Muse, Lehman Brothers.

Yes good afternoon, thanks for taking my question. I guess a couple of questions here; first off, on the Harmony DRAM ramp constraint, is it safe to assume that those customers have gone elsewhere for probe cards, and I guess the question to that is how sticky will those relationships be, and can you get that share back?

So -- this is Igor -- so we Harmony addresses really multiple segments within DRAM. There are, and I assume you're asking specifically about DRAM.

Right.

There are segments in DRAM only FormFactor can address. In those cases where we are not providing Harmony for those, then an alternative would be our previous generation probe cards. It has been our experience so far that when customers can get our product, they definitely prefer our product. That's been our experience always in the case to the Company. And they're not any different for Harmony.

There are 2 other companies that are claiming capabilities for full wafer contactors for the simpler applications in DRAM, where they do not require the number of probes, or do not require the path (inaudible) that we can support. And one of those companies uses 2D advanced technology, and the other one, as we understand, uses manual kind of technology.

So in those cases, the business will go elsewhere. We are confident as we ramp as our lead times will come down, that customers will prefer our product. So we are very, very focused right now. I mean, the enemy right now we see is we see in the mirror so we've just got to execute and we'll do what it takes to execute.

Okay that's helpful; on the Harmony NAND side, you talked about I guess orders now with 2 customers, and I guess for that second customer, can you talk about the magnitude there and whether that's a follow-on type of eval or whether you're making real strides in terms of penetrating that account?

We received significant follow-on orders for 2 customers in NAND, and our activity in NAND again will be subject to how quickly we can get Harmony into high-volume production, high-volume manufacture.

Got you, and last question for me, I guess going back to a prior question, on the slowdown in terms of DRAM revenues for you guys, it looks like it grew 17% for the first 9 months of this year year-over-year. Can you comment given the transition to higher densities, DDR3, 70 nanometer, and lower shrinks, whether you expect in 2008 that you'll grow faster or slower than that level?

As far as -- I'm not sure which 17% you're referring -- but as far as '08, as we did this year; this year in January we came out with overall view of what we plan to do in '07 and we will do again the same basically guidance for '08, but we will do it on the next earnings call. But we will do this again; we will estimate what we feel the market will grow, and we will estimate what we feel FormFactor will grow.

Thank you very much.

Thank you.

Harlan Sur, Morgan Stanley.

Hi, good afternoon; so if we didn't have the constraints here on the Harmony-based products, would the Company be sitting here today with higher bookings in Q3 and higher revenue guidance for Q4?

Yes.

Okay, and as it relates to the manufacturing constraint, I mean is this going to impact potential revenue growth in Q1 and Q2 of next year as well?

We will be working on Harmony manufacturing ramp in Q4 and Q1 most likely, so we will, of course we would like to make it production worth it as soon as possible and high-volume manufacturing worth it, but basically a lot of hard work in Q4 and probably part of Q1.

Okay got it, and I know you spent a lot of time talking about the demand for your Harmony XT products. Igor maybe if you could just talk about, or comment about your PH150XP products and the demand trends there.

So that remains the workhorse for the industry, that remains the workhorse for the majority of the industry. And as I mentioned, for the more demanding applications, that's still the product that the customers buy when we don't deliver Harmony. Although, there are again, some parts of the market where the compelling advantages of a full wafer contactors, in terms of test cost reduction is such that customers will look at alternatives, even if alternatives are not as good as what FormFactor offers. We've synthesized the market very well to these test cost advantages, so what we need to do right now is to get Harmony execution into different gear.

Yes, got it okay, and then one final question for Ron. How many 10% customers in the quarter?

There were three.

Okay, all right, thank you.

[Fanchion], [Elpreida] and Powerchip.

Gary Hsueh, CIBC World Markets.

Hi, there's a lot of things here I don't quite understand. Let me just start with number 1; Ron what gives you the confidence that you can kind of maintain a 20.5% to 21% operating margin in Q3? Why is that not up for grabs? Did you already start to change our inventory valuation methodology here in Q3?

Gary, that's a good question; obviously, we're trying to get things resolved as quick as possible, and given the scheduled call, we wanted to give you the input we had at this point in time. But one thing you need to be aware of is we have spent more time looking at our Q3 results. Obviously, that's what surfaced the issue in the first place, and we spent a lot more time scrutinizing Q3.

So that has given us a degree of comfort that we can still reconfirm our guidance for the third quarter at least in the range that we communicated on the profit side.

Okay, so Ron that sounds like a yes, you already started that more conservative, if I may, inventory valuation methodology in Q3, and so there's less risk here to a restatement in terms of your operating margin guidance for Q3? Is that the right interpretation?

Let me be clear; we're not changing policies. We are validating, in fact, the policies that we've had in place for some time, have been consistently followed. So there's not an issue of changing that policy; we are looking back in '07 and prior periods to assess whether or not that policy was properly applied, and that policy is still in force going forward.

So we have done an intense amount of scrutiny on Q3, and therefore we're comfortable with reaffirming the guidance that we gave for Q3, that we can at least perform on the profit line at that level.

Okay and frankly, I'm a little surprised why you all of a sudden needed to start spending a little bit more in engineering and development to support a ramp here in Harmony, and also in manufacturing startup costs here in Q4. I mean, the Harmony product, correct me if I'm wrong, has been out for at least 2 to 3 quarters since late last year, and unless you were betting against your Harmony product getting commercial attraction, I don't know why all of a sudden it's a surprise and something that you need to start spending for and something that's starting to deviate from your financial target all of a sudden here in Q4. I mean, can you explain that to me?

We -- this is Igor -- we, what we do is we add manufacturing capacity, historically we've added it at a certain point for a new product ramp. And what we're saying is that in this particular case, we decided to add additional capacity ahead of where we would have normally done it. So we made it, and it's not necessarily that, as I mentioned before, there are no 2 ways, we're finding it more difficult to do, and what we're doing is we're adding more, adding more people, more equipment, and more space to solve the problem sooner. And that's both financial and efficient.

Okay.

It's not engineering and development; entire exercise right now is to get this into manufacturing hands, into high-volume manufacturing.

Yes Igor, my point was it was supposed to be in manufacturing high-volume hands earlier in the year, and it got pushed out, it got pushed out, and so I thought the infrastructure was already there for this.

You are correct; we are finding it more difficult to do.

Okay, and my last question is for Ron; I mean, you talk about ramping Singapore, if I got this correct, in the first half of '08, and some increase here in SG&A I've got to believe in the first half of '08, and we just expected 17% to 19% GAAP operating margin guidance here for the first half of '08 as well? Is that kind of a fair sort of assumption to make in our model?

Well first of all, to be clear, we're ramping sales operations in Singapore in the first half of '08. Manufacturing will be ramping in early '09. So it is not the big cost structure of a factory that we're putting in in '08 just to be clear on that.

Right, but I'm just looking at the SG&A line here; I'm seeing upward pressure in SG&A in the first half of '08, and I'm just wondering if that upward pressure on SG&A in the first half of '08 is going to keep a lid on operating margins in the 17% to 19% range.

I see your question.

Yes.

We believe we can operate within our financial model including on the SG&A line, as we move into Singapore. And there will be some incremental costs, but most of the investments we made in Singapore to date are actually adding value. For example, my comment about the engineering resources we've put in place, design engineering in Singapore, etc.

So that'll be managed going forward, and when we talk about sales operations in Singapore, we're talking about some sales management, some which is already in place, and we're talking about order processing and other financial activities, collection, and account receivable management, those sorts of things, that we gear up Singapore for a broader production.

That enables us to capture some of the benefits of intellectual property transfer in Singapore in 2008, and brings our tax rate down for our estimates to the mid-30s in 2008 by making that move. And the big leverage in Singapore will come obviously as we ramp the factory and the tax rates start coming down more materially from there. Does that address your question?

Yes that helps, thanks.

Just another point of clarification Gary because we're not saying we've got anything other than a near-term problem related to a Harmony ramp, and let me be clear in Q1, we believe that we can actually double what we, we plan to actually double our capacity. The fact is that the demand from our customers is so robust that it may take us until the second quarter to catch up. But all that ramping and coming down the cost curve, from now through the first quarter, and the ramp-up volume will all leverage our margin line and help keep us on the model.

Doug Reid, Thomas Weisel Partners.

Thanks for taking my question, and I have a couple. If I just look at the first 9 months of the year, it looks as though the sort portion of DRAM revenue goes to an annual percent, an annual growth rate, and I know you've talked about the general slowdown in DRAM-related revenue growth. But I'm trying to understand what part of that slowdown relates to market share loss in the sort category. So I'm wondering if you could give an estimate of what you think FormFactor's market share is in sort.

We are not, I don't believe in the first 3 quarters we lost market share in sort DRAM. I think if you look at the impact of full wafer contactors, say up to now, that they're related to us being behind with Harmony, that would be a few million dollars in total, right, and it's a very large business.

What we now need to do is just as we are planning right now to make progress at a good clip, and keep our leadership in DRAM. I'm not sure that the magnitude of this full wafer contactor business up to this point, it's not very large part of DRAM business.

Yes Doug, another way to think about it is if you look at our core PH15 DRAM business, well we're very well established with all our customers the capability of our new products, the PH150XP, etc., in the sort category. Right now there are 3 customers who are working on getting some of their capability ramped in qualification phase on full wafer contactors, and we are engaged with all 3 of them.

So it's a, just an emerging market variable and opportunity for us going forward, and we're working the ramp as fast as we can to keep pace, not only with those initial adaptor customers, but be ready for the next ones in the next quarter as they come on board.

I do want to make it really clear, I mean this product, the Harmony product work in a high-volume [infection] environment at multiple customers, both in DRAM and Flash. I visited customers personally, I have customers which they usually don't do, telling me they're getting high yields, which normally, which never is good negotiating strategy with anybody. And so this will be great product.

We are right now, what we're doing is putting a lot of effort into making sure that we pull in the (inaudible) curve for the Company, and I would not be connecting this with any market share loss right now.

Okay, and I realize you don't want to give explicit revenue guidance for 2008, but I'm wondering if you can, based on your understanding of customer technology transitions and your understanding of the seasonality in mobile, help us understand what kind of profile revenue growth might take in 2008; front-half loaded, back-half? What can you help us with?

We really would prefer to do that on the next earnings call. We did it in the beginning of '07, and we'll do it again at the beginning of '08.

And we're in that planning process right now Doug, so we'd like to get that work completed and then share it with you.

And then lastly if I could just a sort of quick final question, could you give an update on your patent litigation? Any changes in strategy there given the recent loss in Korea?

A trial date has been set in Oregon, and that's the patent case against [Ficomn], the Korean company. There has not been yet any date set in [MGC] case, which is in Northern California, and MGC is the Japanese company. We will very, very strongly defend our IP. So we are considering all kinds of avenues to make sure that FormFactor IP is protected.

Okay thank you.

Ladies and gentlemen, as this is all the time we have for Q&A today, I will turn the call back to management for closing remarks.

I want to thank you all for joining us today, and we certainly look forward to seeing you at future conferences and on our next earnings call. Thank you very much.

Ladies and gentlemen, we thank you for your participation in today's conference call. This does conclude the presentation and you may now disconnect. Have a good day.