FormFactor Inc (FORM) 2004 Q3 法說會逐字稿

Good day, everyone, and welcome to the FormFactor third-quarter 2004 earnings conference call. This call is being Web cast live on the Investor Relations page of FormFactor's Web site at www.formfactor.com. (OPERATOR INSTRUCTIONS). I would now like to turn the call over to Ms. Erica Mannion, Investor Relations for FormFactor Inc.. Ms. Mannion, you may begin.

Thank you. Good afternoon, everyone, and thank you for joining us for FormFactor's 2004 third-quarter earnings conference call. Shortly after market closed today, the Company put out a press release announcing its third-quarter and financial results. If you did not receive a copy of the press release, you can obtain one from the Investors section of FormFactor's Web site at www.formfactor.com. You may listen to an audio replay of this conference call by dialing 888-203-1112 and entering reservation number 904083. Also, a Web cast replay will be available in approximately two hours on the Investors section of the Company's Web site.

With me today are Igor Khandros, President and Chief Executive Officer, and Jens Meyerhoff, Chief Operating Officer and Chief Financial Officer. Igor Khandros will begin with an overview of the Company's achievements and progress during the third quarter of 2004. Jens Meyerhoff will provide you with details of the third-quarter financial results and financial guidance for the fourth quarter of 2004. We will then open the call up for questions.

The Company has allocated approximately one hour for today's call. During the Q&A period as a courtesy to those individuals seeking to ask questions, we ask that participants limit themselves to one question and one follow-up question.

Please note that details about the Company's revenues and bookings by market segment, as well as a regional breakdown of its revenues, can be obtained from the Investors section of FormFactor's Web site at www.formfactor.com. This information will remain available until October 22 at 9:00 PM Pacific Daylight time.

During the fourth quarter of 2004, the Company will be presenting at the following conferences. Morgan Stanley Small Cap Executive Conference in Florida, November 8; Lehman Brothers Small Cap Conference in Scottsdale, Arizona, November 18; Lehman Brothers Technology Conference in San Francisco, December 8.

Now I would like to make a brief statement regarding forward-looking remarks that you may hear today on the call. During the course of this call, the Company will make projections and other comments that are forward-looking statements within the meaning of the Federal Securities laws. These statements include a number of risks and uncertainties. In addition, statements regarding design wins and bookings should not be read as projections or predications of future performance. These statements are based on current information and expectations that are inherently subject to change and involve a number of risks and uncertainties. We caution you that actual events or results may differ materially from those in any forward-looking statement due to various factors including, but not limited to, the demand for certain semiconductor devices; the rate at which semiconductor manufacturers make the transition to 110 and sub-110 nanometer technology nodes; the rate at which semiconductor manufacturers expand their 300 millimeter wafer manufacturing capacity; the performance and market acceptance of the Company's new products or technologies; the implementation of volume production of the Company's new products and the Company's ability to efficiently build out and move into its new manufacturing facility; changes in semiconductor manufacturers' test strategies, equipment or processes and the Company's relationship with customers and companies that manufacture semiconductor test equipment.

Additional information concerning factors that could cause actual events or results to differ materially from those in any forward-looking statement is contained in the Company's Form 10-Q for the period ending June 26, 2004 filed with the Securities and Exchange Commission and subsequent SEC filings made by the Company and in the Safe Harbor language in the earnings release that was sent out today. Copies of filings made by the Company with the SEC are available at the Investors section of the Company's Web site. The Company assumes no obligation to update any statements made during today's call, to revise any forward-looking statements or to update the reasons actual results could differ materially from those anticipated in forward-looking statements.

It is now my pleasure to introduce Igor Khandros, President and CEO of FormFactor. Igor?

Good afternoon and thank you for joining FormFactor's third-quarter 2004 earnings call. The third quarter of 2004 marked yet another strong period for FormFactor. Our revenues were up 19 percent sequentially and 97 percent year-over-year. Our net income rose by 11 percent sequentially to 7.5 million.

During the third quarter of 2004, FormFactor still remains heavily focused on the continued increase of our production capacity. We are pleased with the incremental capacity ramp of 8.2 million in products shipped, the strongest experience in the Company's history.

You may recall that we were essentially sold out through the first half of 2004 and had to extend our lead times as a result of limited capacity. This brought strong order flow as our customers started to book further out. It also brought a sharp decline in our turns ratio as our backlog grew to over 13 weeks.

The strong performance of our operations team allowed us to communicate to our customers reduced lead times towards the end of the third quarter as the additional capacity came online. As a result, our customers began to normalize their bookings pattern with a corresponding sequential decline in bookings and an associated increase in turns sales.

The fundamentals in our market appear to remain robust with continued big growth in both Flash and DRAM memory segments, and further tooling cycles that are either in progress or signaled by the designs that we have in-house.

We witnessed continued strength in the DRAM segment. Our DRAM-related bookings increased from 30.4 million in the second quarter to 31.9 million in the third quarter despite the shortened lead times mentioned earlier. The acceptance of our newly released 300 millimeter 4-touchdown page 150 ads platform has been rapid and received 3.8 million of this new product to date.

After a decline of DDR2 bookings during the second quarter, we saw a sharp rise in orders for this architecture. DDR2 bookings increased by 9.5 million in the third quarter to 14.2 million compared to 4.7 million during the second quarter. The additional turns generated by our operations team allowed DDR2-related shipments to more than double from 4.6 million during the second quarter to 9.8 million in the third quarter.

DDR2 design activity maintained the strength from the second quarter. While our DDR2 business is gaining momentum, we do believe that this tooling cycle has not yet emerged in its full strength. The adoption of the DDR2 architecture appears to be moving somewhat slower than what we experienced with prior transitions, suggesting that the bit parity for DDR2 may be reached in the second half of 2005 compared to earlier predictions of Q2 of 2005.

From FormFactor's execution standpoint, this timing gives us an important opportunity to complete the transition into our new more efficient production facility, ahead of the tooling milestone of DDR2, DDR1 bit parity. As we mentioned before, we expect the DDR2 revenue opportunity to exceed the DDR1 cycle by about a factor of 2.5.

Another encouraging data point in our business is the continued ramp of sub-110 nanometer designs. During the second quarter, I reported to you increased design in bookings activity for sub-110 nanometer DRAM devices. In the third quarter, shipments of both 100 and 90 nanometer designs increased from 1.2 million to 3.3 million. 110 nanometer and below related DRAM revenues accounted for 83 percent of our DRAM revenues in Q3 compared to 69 percent in Q2. We expect this trend to continue as sub-110 nanometer bookings increase from 3.7 million in the second quarter to 8.1 million in the third quarter of 2004.

Our TRE technology allows our customers to extend mutual scaling generated by these node shrinks into test as the increased number of die or bits per wafer can be handled with the existing installed base of testers, enabling a significant reduction in the cost of tests.

512 megabit designs resumed growth during the quarter and accounted for 13.1 million of our revenues, up from 7.3 million in the second quarter. It appears that 512 megabit density node is now rapidly moving towards becoming the standard in most PC applications.

Model RAM demand weakened during the third quarter, possibly due to slower demand for digital consumer devices. Revenues declined to 4.5 million in the third quarter, down from 5.4 million in the second quarter. However, year-over-year growth for Mobile RAM continues strong, with revenues up more than a four-fold. The strong rise in DRAM content and portable consumer products driven by VDE imaging continues to be a long-term growth driver for these low-power devices. It is clear now that the use of Mobile RAM in the mobile consumer market has ushered the era of known Good Die manufacturing into our industry, and the projected increase in use of the industries only complete known Good Die used by FormFactor, the S200 high-speed wafer test probed cards and our wafer level burn-in products.

The bringing up of new 300 millimeter wafer capacity is continuing in the industry. 300 millimeter shipments during the third quarter of 2004 continued to increase to $16 million and accounted for 46 percent of our DRAM product shipments, up from 38 percent in the second quarter of 2004.

Our Flash revenues continued to increase from 10.8 million in the second quarter to 12.7 million in the third quarter of 2004 as a result of the significant Flash bookings during the second quarter. With our new factory coming online, we will continue to aggressively pursue opportunities in this fast-growing market as we plan to introduce our industry-leading 300 millimeter 4-touchdown page 150s product into Flash.

The continued ramp of 300 millimeter production drives strong demand for this and other FormFactor high-volume wafer probing solutions in this growing market segment.

Flipchip Logic-based revenues remained soft with 3.4 million in the third quarter, down from 4.2 million in the second quarter. With the stronger than expected execution of our capacity plans, we intend to be more aggressive in this market segment going forward. Our recent design wins in this market segment suggest growth as we move into 2005.

Our overriding priority in 2004 has been the execution on the growing deliveries into customers' mission critical applications and building the manufacturing springboard for future growth. Through the first nine months of 2004, FormFactor has grown 97 percent over the same 2003 period. We can add the challenge of balancing the remarkable growth of our existing manufacturing capacity with this joint execution on the bring up of our redundant new factory.

A week ago we officially opened the new campus and successfully completed our first move. We believe that the new factory is the most advanced (inaudible) manufacturing facility of its kind in the world. The bring up of the new factory continues to be fundamentally on track and should provide some incremental capacity during this fourth quarter. The recent softness in the overall semiconductor market does not appear to have reached FormFactor's core markets, but during the third quarter we have seen an increase in the design activity that has historically signaled to us a potential change in the cycle.

Counterbalancing any potential softness in the end markets will be our growing ability to turn orders in a current quarter, as well as increased business generated by the increased design intake in the third quarter. While the overall impact may possibly be a reduction of our revenue growth rate from somewhat unsustainable levels for our existing production facility, it gives us an important advantage of more predictable transition to our new facility.

As a part of our execution plan this year, we grew headcount by 45 percent during the first three quarters of 2004, and we have have focused on the task of assimilating and training new personnel. I look with proud on the performance of our team. Commitment, dedication and competence of our people manifest themselves not only in a timely on budget execution, but also in the outstanding financial performance in an environment challenged by rapid growth. The pure fact that we effectively generated enough cash throughout this year to so far fully fund our new site investment exemplifies the level of our performance and execution.

As a part of our commitment to sharing with our employees, we implemented a profit-sharing plan in full appreciation for their contributions to FormFactor's growth and financial success. This plan is structured to contribute up to 7.5 percent of our Company's payroll back to our employees and is structured to further bolster our competitiveness in attracting and retaining the best employees. Please note the Vice President and executive officers are not eligible for this program. The implementation cost of this plan was $370,000 after taxes or .8 cents per fully diluted share during the third quarter, an investment that will carry a strong return as we move towards the future challenges and future growth.

I will now hand the call over to Jens Meyerhoff, our Chief Operating Officer and the current Chief Financial Officer, for a more detailed review of our operations and third-quarter financial results.

Thank you, Igor, and good afternoon. We had another strong quarter. I am particularly pleased with our ability to ramp up production output by over 8 million, an unprecedented number for FormFactor. This important achievement allowed us to catch up with our customers' demand, to reduce lead times and to ship products to our customers which are mission critical to them.

Revenues for the third quarter were $51.4 million, an increase of 8.2 million or 19 percent over the second quarter of 2004 and an increase of $25.3 million or 97 percent compared to the same period of last year. Geographically we continued the trend seen during the second quarter with the amount of strength in Asia, especially Taiwan and Korea, but also further growth in Europe. Revenues increased in all regions, except North America, due to weaker logic demand.

During the third quarter, our bookings declines by 12 percent sequentially, but continued with year-over-year growth of 55 percent. In the past we have explained to you why bookings alone are not a very meaningful indicator for FormFactor. However, given the sensitivity to bookings in the semi-cap equipment industry, I will highlight once again the relationship of bookings and turns ratio as it relates to a design driven business like FormFactor.

When we signaled to our customers that capacity was tightening during the first quarter of 2004, our bookings increased sharply in the first quarter of 2004, but our turns ratio declined from 53 percent in the fourth quarter of 2003 to 35 percent in the second quarter of 2004. At the same time, our backlog began to grow from our traditional six to eight weeks to over one quarter. As a matter of fact, we indicated during our last call that most orders booked in Q2 converted to shippable backlog in the third and fourth quarter of 2004.

In the third quarter, we communicated reduced lead times to our customers. The resulting moderate declined lead to a realignment of our backlog to our shortened lead times, but also brought the expected increase in the turns ratio. The third quarter turns ratio increased to 43 percent for the quarter, following the announcement of the reduced lead times. We expect the turns ratio to further normalize during the fourth quarter towards historically higher levels.

Gross margins for the third quarter were 50.1 percent, down from 52.9 percent in the second quarter and up from 48.7 percent in the third quarter of 2003. The decrease in gross margin during the third quarter was driven by non-recurring new site expenses of $1.2 million as per our guidances; new site recurring expenses of $563,000 as we began to depreciate portions of the new site; the manufacturing-related portion of a profit-sharing plan of $349,000; an increase in inventory reserves of $700,000, and the ramp-up costs for our PS (ph) 150s product in the amount of $302,000.

The impact of these factors over the second quarter on gross margin is 4.5 percent, was 1.8 percent related to the new site bring up which continues to remain below budget and prior projections. Productivity in our core operations remained at the strong second-quarter levels.

The fact that our gross margins have remained above 50 percent through the bring up of our new site is quite remarkable. The incremental revenue growth combined with the bring up (inaudible) ability to substantially remain below budget has afforded us a minimal dilution in gross margins to date.

Operating expenses were $13.9 million in the third quarter of 2004, up from $11.9 million in the second quarter of 2004. The increase is a result of increased personnel expenses in R&D, as well as in SG&A. During the third quarter, legal and compliance costs increased to $800,000 over the second quarter of 2004. We spent $500,000 on current litigation, an increase over our initial guidance. Our spending on Sarbanes-Oxley 404 compliance increased significantly during the third quarter to $300,000 and will continue to grow through the remainder of the year.

With that, operating expenses, and excluding stock-based compensation as a percent of sales, were 26 percent in the third quarter, flat from the 26 percent in the second quarter of 2004. Operating income for the third quarter was 23 percent or $11.8 million, an increase of $900,000 over the second quarter of 2004 and an increase of $8.7 million over the third quarter of 2003. Operating income for the third quarter included $609,000 for the amortization of deferred stock-based compensation.

Interest income for the quarter was $635,000, reflecting an average yield of 1.5 percent for primarily tax-exempt commercial paper. The tax rate for the third quarter was 39.1 percent, down from 39.8 percent in the second quarter of 2004 and up from 38.4 percent in the third quarter of 2003.

Net income for the quarter was 7.5 million or 19 cents per fully diluted share compared to $6.8 million or 17 cents per fully diluted share during the second quarter of 2004 and 2.2 million or 6 cents per fully diluted share during the third quarter of 2003.

Cash and marketable securities increased by $6.3 million during the third quarter. Cash flow from operations during the third quarter of 2004 was $9.3 million compared to $8 million in the second quarter of 2004 and $4 million during the same quarter of 2003. We spent $6.6 million in capital expenditures during the third quarter against depreciation and amortization of $1.7 million, and received $2.2 million from the issuance of stock driven by our employee stock purchase plan and employee stock option exercises.

Our DSO decreased to 48 days during the third quarter of 2004 compared to 60 days in the second quarter of 2004. Net inventory increased by $446,000 during the third quarter as a result of the continued ramp of production. The overall result is that year-to-date we have funded the new facility investment as well as our rapid growth during the three quarters without reducing any of our cash reserves.

This brings me to our guidance for the fourth quarter of 2004. We expect revenues for the fourth quarter to be $50 to $53 million. Operating income is expected in the range of 22 to 24 percent. We expect stock-based compensation of $800,000 for the first quarter, and we expect to incur approximately $1.5 to $1.7 million of non-recurring expenses related to the new site and $800,000 of recurring costs such as depreciation and facility costs.

We expect to incur approximately $500,000 of legal expenses and current litigation. We target earnings per share of 18 to 19 cents per fully diluted share.

With that, we will now open the call for questions. Operator?

(OPERATOR INSTRUCTIONS). Jim Covello, Goldman Sachs.

Jens, could you do me a favor? Could you please run through the extraordinary charges in cogs and operating expenses again? You went through all the charges that were in there. Some of those were not extraordinary, though. Could you help me out one more time with that, please?

Certainly, I will do that. What we were referring to were non-recurring new site expenses of $1.2 million, which was consistent with our guidance in the last call. The incremental impact of that is $700,000 or 1.4 percent.

We had new site recurring expenses as we started to incur depreciation, and again the incremental was $200,000. The total amount was $563,000. The manufacturing-related portion of the profit-sharing plan, which is a new plan as you heard from Igor that was implemented during the third quarter, cost us $349,000 in cogs, which is .7 percent incrementally.

I don't mean to interrupt, but that is not extraordinary. That is going to be recurring. (multiple speakers). So I want to focus on the non-recurring charges. (multiple speakers) -- the 1.2 million facility bring up?

Right. So, Jim, it will be an ongoing expense, but it is an incremental cost over the second quarter. That is why I am highlighting it.

Absolutely. I understand that. (multiple speakers) -- we have the 1.2 and then what else?

We have the 1.2 and we have $300,000 of PS (ph) 150s ramp cost as we bring up the yield on that productline, which will grow into volume production as we move forward.

Okay, okay. And then relative to the guidance of 18 to 19 cents, does that include or exclude the facility, the million or 2 or so in facility charges bring up and then any amortization expense?

So we always guide a GAAP number. This includes all expenses.

Great. Maybe if I can get one more for now. Going forward, how should we think about bookings versus revenues? Should we always expect bookings to be a little bit below revenue because there is always going to be some turns business in the quarter?

Longer-term obviously you have to have a balanced scenario of bookings and revenues. But if you look right now at our revenue guidance of 50 to 53 percent, and you do the math over an increase of the turns ratio, I think we will get a resulting answer on the bookings side. You know we are not guiding particularly to bookings.

Right. Okay. Thanks very much.

Ed White, Lehman Brothers.

Two questions. First, how does the turns activity look for the fourth quarter? You mentioned it was strong in the third quarter. As we look out a little bit to the future, what are your thoughts on that?

Historically we have seen a turns ratio of 50 to 55 percent, so we expect as we are normalizing right now that we will get over time back to these historical levels.

Okay. If you look at the design activity, you mentioned the design activity more robust. Can you talk about what the design activity is telling you now about where you are likely to be as you looked at the quarters ahead? Do you think that is going to slow down, or do you think that is going to remain robust? What are your thoughts on that?

Okay. I'm going to answer this a little bit more broadly, but usually what we have seen, as we see potential shift in the overall cycle, that the overall amount of designs is increasing in the industry, which drives additional business for us. So I think we will probably not be able to predict for you where we are exactly in the cycle, but we have at least seen as we indicated in our script we have seen the first signs of increased design activity, and especially for memory, you see that there are some strong tooling cycles in front of us as we rapidly appear to be moving to a sub-110 nanometer and you also see that despite a potential slower DDR2 adoption, our business in that architecture has increased significantly.

I will answer that these are unusual times in a way that there are several architectures coexisting and also several processed nodes coexisting I think in an unprecedented way, right? So that drives the proliferation of designs. So we have to be always very very carefully analyzing this data. All in all it is very positive for us actually.

Okay. Finally one quick one. The 700,000 inventory reserve that you mentioned, what was that related to?

This is something actually historically we have seen throughout the third quarter where we see a slight rise in inventory reserves as all our customers appear to fine-tune their mix towards the Christmas business, and as you know, we are purchasing some material against a forecast and the accuracy of the forecast determines the use of those custom materials.

Okay, great. Thank you.

Mark Fitzgerald, Bank of America.

You guys haven't had a flat quarter since 2000 in terms of the fourth quarter here. Are you attending this basically to the delays that are going on in DDR2? Is that the core issue here?

As we mentioned in the script, Mark, we had three quarters of constraint capacity, and we are working very very hard to bring up new facility while squeezing additional capacity out of the existing facility. So that is really the underlying reason for what is going on.

As I mentioned, we expect some incremental capacity from the new facility this quarter, and you will see us communicating to customers the shorter lead times as we have done in the end of Q3 and going after business a lot more aggressively.

What we have done during the year is basically focusing on those applications were there is no alternative to FormFactor. Right? And working again very very hard and diligently on bringing up additional capacity in the existing facility. So it is interplay of those factors is what really drives it more so than industry factors in general.

So you are saying you basically don't have the capacity to ship at higher levels?

No, Mark. I think what we are saying is so that even if we had had even sharper rise in DDR2 bookings, at that point in time you would have seen us being strong straight back into a capacity constrained situation. So the slight delay that we are seeing, the potential delay of one quarter on DDR2, is actually an opportunity for us to continue to build up the capacity in our new site, and as Igor mentioned, we are going to see first incremental capacity in Q4, which we will continue to grow into Q1. And this will fundamentally give us a much better ability to respond to that next strong growth leg that we're going to see out of DDR2.

Okay. I think I understand that, but I'm not connecting the dots here. Because simply if you look historically, you've had a very very strong December quarter, and you are guiding flat at this point. So it's either you cannot supply it and the demand is there, or the demand is not there. So I am not quite sure given all that you just said --?

At this point in time, as we mentioned to you, there is probably -- supply and demand with the current outlook are matching at this point in time. Yes? However, if there would have been a much stronger pull on DDR2, then capacity would have become immediately again the constraining factor.

So as I mentioned in my comments, Mark, on Flash and on Logic, we're getting to the point -- looking at our execution on the existing site, looking at our execution on the new site. We're getting to the point where we will more aggressively go after business outside of this mission critical requirement of customers.

Okay. But bottom line I can conclude that there was a problem here in DDR2 in terms of the fourth-quarter demand, it is a bit weaker than what you would expect seasonally. Is that a fair conclusion?

I think in my comments and in Jens' comments historically we have seen maybe faster trend in architectural traditions, and as we plot the trend for DDR2, it looks like it is probably a quarter or so later than before. Right? That is historical.

And is that across the spectrum of customers that are going into DDR2?

As we have mentioned on earlier calls while we will never get into customer specifics, even here there are some companies that are more aggressive on our early adopters and are tooling more aggressively than others.

Mark Bachman, Pacific Crest Securities.

Your comments today, they have been absent of any mention of your wafer level burning card. Can you tell me how that is going with your beta customers, and are you still on track to get meaningful revenues here in Q4?

Okay. Thanks for asking that question. It is a good point. The last time we talked we were in the status of two customer engagements. At this point in time, we have increased this to five customer engagements. We booked a $1.4 million to date on that product, and we shipped 1 million. And obviously these new added customers are just in the process of adopting this product. So we are seeing a continued acceptance of the product as we move into '05.

So what are your expectations then for Q4 as compared to your Q3 results here?

Well, we have never given guidance down to the product line level, and I am probably going to stay away from this. But we will give you an update as we move into this on our next call.

Can you talk quantitatively rather than qualitatively then?

Well, qualitatively I think what we are seeing across the board is that requirements for known Good Die testing is really taking hold in the industry. And the leading adapting applications for those of Mobile RAMS. But you will see that trend. That trend actually is very encouraging to us, and I believe it is ahead of our expectations.

So what you will see is a long-term trend of companies doing full testing and burn in on a wafer over the next several years. We are just seeing the beginning of it.

Okay. These revenues here, did these come from one or two customers?

As I mentioned to you before, we have two early adopters that are further ahead, and I just added to you three new customers. So your qualitative assessment question is, so you more than doubled the number of engagements.

And then looking at your geographic breakdown here, you had pretty strong sales this time from Europe. Was there a reoccurring 10 percent customer there and who fell off if that happened?

(inaudible) answer this a little bit more broadly, because we have not filed the Q yet and we stay away from customer specifics. But in Europe we are seeing quite strong DRAM demand, and we have also continued to see growth in Flash.

Kevin Vassily, Susquehanna Financial Group.

Good afternoon. I was wondering if you guys could talk a little bit about any kind of difference between the cost of tests for the, maybe what you would turn the current configurations of DDR2, and the cost of tests that will be associated with DDR2? Is there any kind of gap, and what kind of impact do you see or do you think you guys could have that could favorably impact cost of DDR2 tests?

Yes, that is a good question, Kevin. So the DDR2, the fundamental difference is it is a low-power low voltage architecture. Therefore, when you say testing margins, you now have a lot less margin to play with.

So we have seen and expect to continue to be seeing increased yield, for example, when customers use our probe cards in DDR2 application. Of course, the roadmap for performance for DDR2 also is a lot more aggressive than DDR1. As you know, frequency levels are going up, and there the signal integrity and design tools that we have will make a significant difference.

Also, DDR2 architecture has a significantly -- it varies from customer to customer. It has a significantly higher amount of paths that customers probe which drives the overall number of contacts of MicroSpring on a probe card up. It gives us significant competitive advantage and it plays to our strength. And, of course, the pictures will be finer for DDR2.

So there are many reasons why DDR2 testing is significantly more challenging. It will be more challenging also for tester companies because, of course, frequencies are going up, right? And you need higher signal integrity both in wafer tests and still used today packaged tests.

Maybe just kind of an extension on that question. When you're talking to customers and engaged in the sales process, as you look at DDR2, is the discussion wrapped around the entire test and assembly cost, or is it just focused on the test portion for you guys? Kind of the basis of the question is some of the feedback among other things that we have heard regarding the delay in the ramp of DDR2 has to do with the package costs as well being somewhat at least near-term prohibitive and making the transition not all that attractive. And I was just curious how that might have an impact or what kind of impact you guys would have in that discussion if, in fact, everyone is looking at total cost of test and package versus just tests?

Yes. For us the prevailing priority always is that we are never bottlenecked through any kind of architecture transition or any transition (inaudible). We make sure that we line up our roadmaps with customers and end-users to make sure that we are not the bottleneck. So in that sense I think we are succeeding.

The second comment is more of a general nature, and that is every time there is an architectural transition in DRAMs, due to the nature of the industry and due to the nature of who dictates such things in the industry, there are different rumors flying around and all that. But these transitions do take place. Even previously we had much more difficult architectural transitions that took place. So these transitions take place despite different jockeying for position from public relations standpoint by various companies.

Packaging does appear to be a significant right now challenge for DDR2 transition. But companies will figure out how to deal with that and bring it down. We, of course, believe that in the future the best way to do all these things is just use known Good Die. Then you don't need to deal with the problem. So that is what we see a few years from now, and the industry will learn that lesson and the industry will understand that if you want -- and you need to transition to new architecture faster and faster. In order to do that, I think known Good Die in the future is the way to go, and we, of course, will be bringing our products to enable that.

Mehdi Hosseini, Friedman Billings Ramsey.

Just a follow-up to the bookings mix by the end market. We talked about a DRAM and a DDR2 group ramp could be delivered one quarter. But what about the Flash and Logic? What is going on there? Why is the booking down? And specifically with the Flipchip Logic, the PC data is coming up pretty strong, so I would have expected at least some of the seasonality would impact you.

Also as a follow-up, if you could provide some update on the litigation with your competitor, eKorea. Thank you.

Okay, I will take your first question on the bookings by segment. So outside of DRAM you're asking about Flash. All our Flash bookings were around $10 million, $10.5 million, which is down from last quarter by $4.7 million as you pointed out.

But at the same point in time I think we should remember that prior to that we were running our bookings level for 3 to 5 million, and we went through a very strong surge in bookings during the second quarter in Flash and we are maintaining overall a very very high level of roughly 77 percent year-over-year growth.

So the outlook on Flash remains positive for us. The transition to high-parallelism test in Flash is on track and obviously here in place into our core competency. But also the overall bit growth outlook for Flash I think remains quite positive, even though we have been some moderate I think price adjustment in the end market. But usually that price pressure drives accelerated adoption of our product because we are reducing the cost of tests.

Does that explain the lower book-to-bill for Flash compared to DRAM?

So again on the book-to-bill, we shipped a lot of Flash product as a result of the bookings we have in Q2. Right? So you had this very strong surge. We went from in round numbers 5 million to 15 million from Flash bookings in Q2, and now we dropped back slightly towards 10 million, which is still a very very high run-rate for us. And I expect Flash to continue to drive growth for us as we move into '05.

Okay. And on Flipchip Logic?

So on the Flipchip Logic side as we indicated in earlier calls, we actually during the times of capacities constraint declined some of the business there as we did not see it as mission critical as competitors were able to offer some solution to our customers. And so at this point in time, you see the resulting decline in business is still with us, but our design activity in the Logic arena has accelerated because we started to change our position on this market during the third quarter. And as we move again into '05 you should start to see that business to pick up.

And I will just add to that, during the this year one of the bigger challenges for us has been to go to customers and have very frank straightforward discussions about where we are and that we are capacity constrained and that the (inaudible) capacity application and situations where there are no alternatives.

In some cases, the alternatives may not be optimum or where applications you know design requirements are simpler than in more difficult cases. Even if customers needed to do compromises, we had to have those discussions.

As the new site is coming onstream, we will go back and have new discussions with customers about new products and new capabilities, and we will be more aggressive in market segments, including Logic.

You also asked about litigation. As you know, we do not comment on ongoing litigation as you will understand, and as soon as there are disclosable comments we need to make, we will make them. But it just going its course.

And what is the course? Could you just remind us where we are?

The recent litigation that we announced early in the year, and we are diligently working on it.

So you know that these litigations, right, are not taking a quarter or two. These things take a long course. And I think in many statement I have made we should expect this to last a year or potentially longer.

Timm Schultze-Melander, Morgan Stanley.

This is Ben calling for Tim here. Can you comment on the average selling price for the quarter and what is your expectation of the ASP for the December quarter?

Okay, I will take that question. Obviously we have never disclose and will never disclose our average selling prices. I think I explained that on prior calls. But from a qualitative point of view, I would say you the development was positive in the third quarter.

Okay. Thanks.

Peter Wright, CIBC World Markets.

Good afternoon. My first couple of questions are more industry broad and then a couple of specifics. Igor, I was hoping you would be able to comment a follow-up to Ed question on design activity. How did that track through the quarter? Did you see a pickup towards the end, and if you could speak to the momentum that you see through the seasonally weak Q4?

Well, we had pretty strong design activities throughout Q3. We worked hard to work on those designs, so that was not back quarter loaded or anything like that. We just saw a very strong activity as Jens mentioned earlier. Normally it is an indicator of some industrywide -- sometimes it points to some cycle inflection points in the industry.

But as I also mentioned before in the call, these are extraordinary times where you have companies shipping 130 -- 110 -- 190 nanometer nodes where multiple design densities coexist, where there is coexistence of 300 and 200 millimeter capacity, and you have an ongoing DDR1 to DDR2 for example transition in architecture and similar transitions are happening in Flash. All of those intersections, all of those different names I mentioned, they drive new designs and new orders for probe cards potentially.

So again it is something that we are watching very carefully, but it is very hard to definitively state what it means in terms of industry cycle or if it just the fact that design activity will be going up and will be benefiting.

So is it fair to assume that design activity is fairly linear through the quarter?

Yes, it has been strong through the quarter. (multiple speakers)

Also, Peter, I want to add maybe two comments on this. So the design activity is highly customer specific, right? So you're getting these wave patterns of designs that are coming through for the Company. Another thing that drives design intensity right now is that more and more of our product goes into consumer applications where you have a lot faster design cycles. That is true for Flash and Mobile RAM as well.

Great. Looking a little further out, it might be premature to say, but Independent Research House has spoken to DDR3 adopting Flipchip packaging, and I was wondering if you have heard anything (multiple speakers) on this?

That is a good question, Peter. DDR2 is going to be just a fascinating architecture because it is going to take DDR2 features to the extreme in terms of frequency and performance, and it will incorporate in its Mobile RAM low-power features. Right? So you will have low voltage, and you will have low power that is achieved through means like refresh, rate controls and all that. And that is something we are packaging. I don't believe it has been finalized, but what you mentioned, Flipchip packaging is probably looked at. But I do not believe that that is finalized yet.

For us it just will be yet a very very exciting transition because people will just take DDR3 and use it in a lot of consumer applications that will afford not only low-power now but extreme performance. And for us, it's just going to be a huge business driver in the future.

And finally of the dozen or so cards that you have sold, 150s cards that you sold, how many customers are currently ordering those?

Okay. At this point in time, we have got one beta customer an adoption, which actually is a fairly broad constellation of adoption.

Bill Lu, Piper Jaffray.

Good afternoon. A couple of questions. One is in the third quarter you added about $8.2 million in capacity. Was that all in the old factory?

That is correct.

And so I guess it is safe to assume that in the fourth quarter your overall capacity is going to be somewhere north of the third quarter. Can you quantify that?

Okay. So we have never really given a forecast on capacity as a whole. But what I probably will tell you at this point in time is that the existing facility, that this last increment, is at capacity, and all incremental capacity is going to come up in the new site. And that capacity is going to come up in two stages, either through incremental equipment that is already in the new site and in the process of being qualified, and then it starts to move towards a complete production of probe tests and then probe cards.

Okay. So have you shipped anything for production out of the factory to date?

We have not yet shipped a product out of the new factory to date. That would be premature and has never been our plan. We are expecting to get incremental capacity utilization out of the factory in Q4. Complete product shipment is an '05 event.

Okay, perfect. And a couple of follow-up questions. First of all, I understand what you're saying on DDR2, but in the fourth quarter, as you look at some of these non-mission critical applications in Flash and Logic, is it possible that you can turn these into revenues in the fourth quarter, or is that more of an '05 event?

So obviously with respect to our ability now to create again much higher turns, we can be a lot more opportunistic in those applications and opportunities.

Okay. So if we see upside in the fourth quarter, it will likely come more from Flash and Logic then, right?

Turns -- we cannot generate turns in all of the three market segments we're serving.

Okay and then you talked about some new customers at wafer level burn in. Are these all DRAM customers?

That is correct.

And then one last question for Igor. Igor talked about semi design wins in Flipchip in the quarter. Can you just elaborate on that?

We are seeing exciting opportunities outside the microprocessors of chipsets evolving.

Can I assume that these are sort of high-end Logic applications?

These are high-end applications, yes. Very high-performance, very high requirements, yes.

Doug Reid, Thomas Weisel Partners.

Most of my questions have been answered, but I was wondering if you could give us an update on the CFO search?

Yes. So this has been a difficult search for us. We are looking for a partner and a participant on the management team to scale the company to a much larger size. And, therefore, we are just not going to compromise. We have seen confident good candidates, but the moment we see somebody who indeed fits the bill of the inner part of the team that takes this company to a much larger size, we will snatch for him immediately.

Okay.

Bill Grissom (ph), Peregrine.

I just wanted to follow-up a little bit on a previous question. I think you mentioned in prior calls or re categorized that you were deliberately holding back and aggressively pursuing Flipchip opportunities and that maybe they were aware Flash was a year prior. Can you help me understand the dynamic of reramping or ramping the Flipchip sales efforts, and how -- the timeframe it takes for those to flow through into bookings and revenues?

Okay, so I will take a shot at that. The one thing we have done is we have communicated through this capacity constrained times with our customers. We have done that in a very upfront way. Right? So we totally gave people a very true picture of the situation so that they could plan for what we consider non-mission critical for other alternatives.

Obviously our product provides benefits with respect to better yield, with respect to better upside in these applications. And I am confident that given that we have this upfront communication with these customers, we have an easy re-entry into these opportunities because we did not create user disappointment or frustration with the customers as we went through this process.

Can you help me understand what sales cycles are like there?

Could you repeat that, please?

Can you help me understand what sales cycles are like there?

Okay. Those sales cycles are obviously for existing customer engagements or very rapid lead times of orders which you could measure in weeks. New customer engagements obviously on the logic slide especially when you talk about complex silicon take longer, but I think Igor just indicated in an earlier answer that we have ongoing design activity here outside of chipsets and microprocessors, which means a large portion of the penetration time has expired at this point in time. So those opportunities should emerge into revenue over the next probably three to six months.

That is helpful. Thank you.

Scott Jordan, Glenn Capital Management.

Actually it's Steve Ralston. Can you elaborate a little bit more on the progress of your 300 millimeter 4-touchdown product in Flash, where you are and where you are expected to be in six months?

Okay, so at this point in time, Steve, we introduced this product into DRAM into the DDR2 RAM. And so all the revenues of $2.8 million that we mentioned are all DRAM-related at this point in time. We decided to do that because we wanted to scale the platform for the highest pin count and highest design complexity. With that now it is fairly easy for us to scale back, and we are in active discussions with multiple customers for this product opportunity as we go through Q4 into Q1.

And can you comment a little bit about the interest level and importance to your customers there since you did call it out?

So the importance and I think the benefit of this product on the Flash slide is similar to what we have seen on what we see on the DRAM side. Maybe the only caveat is that in Flash you have less test and less probe complexity to deal with. So you're dealing with a significantly lower pin count, but the economics of reducing touchdown from 300 millimeter probe are similar.

Mehdi Hosseini, Friedman Billings Ramsey.

Just a follow-up question to the burn in test question that was asked earlier. As we look into 2005, especially on the back end of 2005, would you expect the revenue contribution from burn in cash or some of these new applications to help generate at least 10 percent of the royal revenue, at least become substantial? And if you can not quantify, just help us to understand how you were able to penetrate these (inaudible) market? Thank you.

This is Igor. You know what we're seeing is increasing demand in what we call known Good Die testing, and as historically again we mentioned that that has been driven by Mobile RAM requirements. Mobile RAM is projected within a few years to reach 20 percent of DRAM industry. So we certainly today as we're looking at 2005, yes, we expect this to ram through 2005 to higher levels. I will not sit here and tell you percentages, but you should see that increasing.

And what it is is a combination of two products or two families of products. One is to be able to do high-speed testing on a wafer. Our announced product is S200, which allows you to test one wafer DRAMs or Flash at high-power (inaudible) adopted 200 megahertz frequency. We are working on products that will go beyond that. And on the wafer level burn in side, it is the ability to perform wafer level burn in, of course, on a wafer at that combination of the two and associated applications support.

It is not just the point of selling it to customers, but you have to go install such things. You have to work with customers to make it work because we are finding that we probably know more about it now than anybody. That is what is going to drive overall business and it will rise in 2005.

Just to follow-up on the known Good Die issue product, application Mobile RAM, would it be fair to say that much of the packaging is done at subcontractors, and to that extent, would you expect those packaging facilities or companies to become major customers in '05?

Well, most of those go into multichip packages, and so the test most likely will always be performed close to the wafer fabrication side.

So what happens is that IDMs or RDRAM or Flash foundries are going to perform the testing next to the front end. It is more more important to perform it right next to the front end, and then they will ship wafers with devices that are fully tested to these sub-counts, and sub-counts basically will just dice and do the rest and, as Jens mentioned, goes into stacked packages, goes into enter really rapidly proliferating all kinds of multichip packages, where inside one package or inside one stack you have different devices. People are looking more and more into executing digital analog mixes that way. So I think it's going to get exciting for us even beyond memory in the future.

So the net is, if I understand you correctly -- the conclusion is that you don't expect a significant shortage to be migrating to subcontractors or actually staying with IDMs, correct?

The majority of it will. The rest from independent testing houses that could do -- this is very sophisticated testing we are talking about. This is beyond plumbing, right? This is extremely sophisticated test programs, very much guarded by IDMs, and it is all now one integrated really manufacturing adjacent to front end. But there are some test houses out there that are developing expertise in that, yes.

And at this time, I would like to turn the conference back over to Mr. Igor Khandros for any additional or closing comments.

As always we appreciate your interest in FormFactor, and thank you very much and have a great afternoon.

This does conclude today's conference. We thank you for your participation. You may now disconnect.