PDF Solutions Inc (PDFS) 2004 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to PDF Solutions, Incorporated conference call to discuss its financial results for the third fiscal quarter ended Thursday, September the 30th, 2004. At this time, all participants are in listen-only mode. Later we will conduct a question-and-answer session for which instructions will be given at that time. (OPERATOR INSTRUCTIONS.)

As a reminder, this conference is being recorded. If you have not yet received a copy of the corresponding press releases, they have been posted on PDF's website at www.pdf.com.

Some of the statements that will be made in the course of this conference are forward looking, including statements regarding PDF's future financial results and performance, growth rates and demands for its solutions. PDF's actual results could differ materially. You should refer to the section entitled "Factors Which May Affect Future Results," on pages 25 to 31, on PDF's annual report Form 10-K for the fiscal year ended December 31st, 2003, and similar disclosures in subsequent SEC filings.

The forward-looking statements and risks stated in the conference call are based on information available to PDF today. PDF assumes no obligations to update them.

Now I'd like to introduce Mr. John Kibarian, PDF President and Chief Executive Officer, and Steve Melman, PDF's Vice President, Financial and Administration, and Chief Financial Officer. Mr. Kibarian, you may go ahead.

Thank you, and welcome everyone. For the third quarter of 2004, we are reporting record total revenue of $16.5 million in non-GAAP earnings of 6 cents per share. Both our revenue and our non-GAAP earnings per share results are the top of the ranges we provided in July.

Gain in share results for the quarter were also in line with our prior expectations.

In a few minutes, I will turn the call over to Steve for more detail on our financial results. But first I have some comments about our business.

PDF Solutions' footprint in the industry continues to grow. During the last quarter we finalized definitive contracts for two engagements with new customers that began in the second quarter, and started another new engagement with existing customers. These additional engagements, all at 90 nanometers, further solidify the leadership of our Integrated Yield Ramp solution for ramping nanometer processes.

A significant factor in our customers' decision to purchase an Integrated Yield Ramp from PDF Solutions is the process characterization technology we provide. Process characterization is the critical component to any yield ramp in every effective DFM solution. Process characterization, in simple terms, is the test chip data required for yield models necessary to make process or design changes that improve yield.

Our solution stands apart because it delivers silicon-proven yield models based on process characterization. By providing CV test chips that statistically characterize the process and design interactions, we enable our customers to accurately understand yield-loss mechanisms and quickly validate solutions.

Further, our solution works with key tools engineers need to ramp nanometer products. Through our partnership with Applied Materials, we integrated our CV test chips with the SEMVision review system. Our customers are able to automatically localize the failures seen in our CV test chips with Applied's tools to more quickly identify the reasons for failures. Setting up inspection and review for a test chip can be time consuming. Using our systems, PDF Solutions customers save time from having to do that work themselves.

The yield models generated by our CV test chips for a process engineer can also be used by the design engineer for manufacturability or DFM. During the quarter, we announced a joint development agreement with Virage Logic to create pDfx-Compliance ASAP logic libraries. Customers running PDF Solutions CV test chips can use the models in the design flow supported by Virage. Our market strategy for pDfx is to have EDA and IP companies' products use PDF Solutions yield models to make the designs created with their tools in libraries more manufacturable. This drives fab usage of PDF Solutions's yield ramp infrastructure, since it is the way to generate data for PDF Solutions yield models. We will continue to work with other EDA and IP providers to deliver PDF Solutions silicon proven models of nanometer process design interactions industry-wide.

Silicon proven process characterization, integrated with industry-centered design and manufacturing tools benefits PDF Solutions' customers in many ways. Our customers get feedback from their first batch of CV test chip wafers produced, accelerating their yield ramp. Our newest customers using our technology during the quarter have already seen the benefits from their first lot of wafers. Many of our customers use the localization feature in Applied Materials SEMVision to quickly localize failures on our CV test chips. Finally, during the quarter a yield ramp customer began to use pDfx, bundled with their EDA tools supplier tool, to create their next chip design, leveraging the yield models derived from that yield ramp. The overall benefit is quicker time to market and higher yields; in other words, a successful ramp.

In closing, when our customers benefit, we benefit. Like many technology companies, we believe our solution brings tremendous value to our customers. Unlike any company in our space, we continue to work with our quarters through the engagement to ensure they realize value, and then we get paid based on the value our customers capture. Since PDF Solutions shares the risk of project success, and since our solutions are field proven, our customers are increasingly willing to apply our technology in the core part of their business, their yield ramps. Our gain-share model appeals to the people and our customers who own the bottom line. It standardizes the payment on one universally regarded benefit, higher yields.

Now we'll turn the call over to Steve, who will discuss the details of our financial results for the third quarter, and provide updated projections for the fourth quarter of 2004, and guidance for the first quarter of 2005. Steve?

Thank you, John, and good afternoon to everyone. First, let me again state that this presentation and our press releases issued earlier today include reference to certain non-GAAP financial measures. The press releases contain a reconciliation of such measures to the most directly comparable GAAP measures, and you may access the press releases and reconciliations in the investor section of our website, located at www.pdf.com.

Revenue for the third quarter ending September 30th, 2004, totaled $16.5 million, a record for PDF, and our seventh quarter of sequential revenue increases. This represents an increase of 46 percent, compared to the third quarter of last year, and an increase of 8 percent sequentially from last quarter. These results were just inside the top of the range provided in July during our last conference call. The increase from the third quarter of last year and last quarter were due to increases in both design-to-silicon-yield solutions and gain share. Design-to-silicon-yield solutions revenue totaled $14 million for the third quarter, a 55 percent increase from the comparable period last year, and an increase of 4 percent from last quarter. The increases from last year and last quarter were the result of greater demand for PDF products and services as 17 engagements from 12 customers each contributed over $250,000 in revenue during the quarter.

As John mentioned, we started a new yield improvement engagement during the quarter at an existing customer, giving us a total of eight new engagements over the nine months of 2004.

Gain-share revenue for the third quarter generated from four customers and five contracts totaled approximately $2.4 million, a 6 percent increase versus the comparable period last year, and an increase of 46 percent from last quarter. Gain share for the third quarter was in the middle of the range provided in July.

We had expected another contract to enter it's gain-share period during the quarter, however, the measurement date based on customer production schedules slipped to Q4.

Gross margins for the third quarter, excluding amortization of core technology, was 64 percent of total revenue, a decrease from 67 percent during the third quarter of last year, and a decrease of less than one half percent from the second quarter of 2004. The decrease in margins from last year was primarily the result of increased travel expenses, recruiting and relocation expenses associated with the expansion of our service personnel in Japan, and higher personnel related expenses. The slight decrease from last quarter was the result of the increase in the aforementioned expenses mostly offset by margins generated from increased gain share.

Total operating expenses before amortization of stock-based compensation and acquired intangible assets were $8.7 million for the quarter, up approximately $1.2 million or 16 percent from the third quarter of 2003, while increasing approximately $297,000 or 4 percent from last quarter.

Research and development expenses for the third quarter increased approximately $519,000 or 12 percent from $4.4 million in the third quarter of 2003, primarily due to increases in personnel related expenses in a larger development organization, primarily the result of our acquisition of IDS in September of 2003, and higher outside technical services expense.

The sequential increase in research and development expenses of approximately $119,000 or 2 percent from the second quarter of 2004 were primarily the result of annual pay increases and the use of outside technical services, partially offset by the continued shift of resources from engineering to client services to accommodate the increase in business activity.

Selling, general and administrative expenses were $3.8 million in the third quarter of 2004, up approximately $695,000, or 22 percent from the third quarter of 2003. This increase was primarily due to increases in sales commissions, legal expenses and financial services, the majority of which were attributable to new business activities.

SG&A expenses compared to the second quarter of 2004, a quarter which included the reversal of excess allowance to doubtful accounts, increased approximately $179,000 or 4 percent, primarily the result of increases in personnel related expenses, including employee sales commissions and outside financial services being more than offset by a reduction in legal expenses and outside sales commissions.

Reiterating the statement made in our press release, in addition to using GAAP results in evaluating PDF's business, management also believes it's useful to measure results using a non-GAAP measure of net income or loss, which excludes amortization of stock-based compensation and acquired intangible assets and the write-offs of in-process research and engineering. Non-GAAP net income for the third quarter ending September 30th totaled approximately $1.5 million, or 6 cents per share. This compares with non-GAAP net income of approximately $332,000, or 1 cent per share, for the comparable period last year, and non-GAAP net income of approximately $1 million or 4 cents per share during the second quarter of 2004. These third quarter results were at the top of the range we provided in July.

On a GAAP basis, including amortization of stock-based compensation and acquired intangible assets, I'm pleased to report net income for the quarter of approximately $135,000 or 1 cent per share.

Turning to our balance sheet at September 30th, total cash increased approximately $467,000 during the quarter to $42.9 million. Operating activities generated $1.2 million in cash during the quarter, and $5.1 million year to date. Capital expenditures for the quarter totaled approximately $500,000, and repurchases under our stock buy-back program used $1.9 million to repurchase another 198,279 shares at an average price of $9.40.

We have now repurchased 505,579 shares for $4.8 million of the $10 million authorized by our board of directors.

Accounts receivable increased approximately $3.7 million to $15.5 million. RA did slip a bit during this quarter as approximately $1.5 million was past due at September 30th, 2004. However, as of this morning, 74 percent of those past-due receivables have been collected, and in addition almost 50 percent of outstanding bill to accounts receivables at September 30th has been collected. Bottom line we remain quite comfortable with our receivable position.

Before I turn to guidance, I will state again that some of the statements made in the course of this conference call, including the ones that we are about to make with respect to Q4 2004 and Q1 2005, are forward looking. These statements include expectations about our future financial results and performance, growth rates, the success of any business objectives, product and service features and introductions, client products and the demand for PDF design-to-silicon-yield solutions. PDF's actual results could differ materially. You should refer to our current SEC filings and understand that the forward-looking statements made during this conference call are based upon information available to PDF today. We assume no obligation to update them.

Now for the fourth quarter of 2004, we reiterate the guidance we provided in our outlook press release earlier today. Revenue is expected in the range of $17.7 to $18.5 million, no change from the guidance we provided in July. Non-GAAP earnings per share is expected in the range of 8 to 10 cents, also no change from our guidance provided in July. And we are also providing guidance for third quarter gain share in a range of $2.8 to 3.2 million, an approximate 16 to 33 percent increase from the third quarter.

Reiterating the guidance for the first quarter of 2005, which we provided earlier today, revenue is projected in the range of $18.0 to $18.8 million, with non-GAAP earnings per share in the range of 9 to 11 cents.

While we believe the bulk of our customer base with its consumer-centric end products will stay the course with regards to their project plans, this guidance does not ignore the negative undercurrents in the semiconductor industry today. As such, while we see broad opportunity in our install base and add new customers, I characterize this guidance as representing cautious optimism that the effect of what appears to be the beginning of a supply-driven downturn will have modest impact on our customers and our financial results.

With that I'd like to turn the call back over to the operator to open the floor for questions.

(OPERATOR INSTRUCTIONS.)

Garo Toomajanian with RBC Capital Markets.

I want to go into that last comment you just made, Steve, about supply-driven downturn. What I'd like to explore maybe is what kind of impact you may have seen thus far from slowing semiconductor growth expectations, and what that really means for customers in their desire indeed to ramp new technology more quickly or more slowly.

Garo, this is John. I'll take that question. During the quarter we had a number of both yield ramp IYRs and dataPOWER deals we were tracking, and actually pDfx deals. All of them closed last quarter, and some of them with companies that for dataPOWER that were expected for the industry that we had expected potentially to be affected. Obviously everyone else has talked about it. Everything that we thought was going to close in Q3 actually did close in Q3. We have a number of discussions going on with customers across our product family from yield ramps to dataPOWER to pDfx for this quarter and again for next quarter, and we do see all those tracking rather well. I think you might have noticed that folks talk about the 90-nanometer node. The capacity is still full and the demand for the nanometer technologies is still very good. So I think Steve's use of the word cautious optimism was just that. As we look at our business, it's been very good, and nothing that we saw in Q3 pushed out, but certainly we're aware of what other people are reporting in the general ecosystem, and so we remain somewhat cautious about what's going to happen for Q1 of next year and the choice for guidance.

Okay, I forget if it was you, John, or Steve, that mentioned that there was one gain-share contract that you expected to start contributing in Q3 that did not. Has that started to contributed yet in Q4?

Yes, we're very confident that we will see our expected results from that engagement this quarter.

Okay, but has it started to contribute yet, or is it too soon to tell?

It's contributing. We have to dot the I's and cross the T's though.

Okay, and on a sequential basis going into Q1 of next year, would you expect gain share to grow sequentially?

That's what Steve's guidance was.

Yeah, we provided the 2.8 to 3.2. That's up --

No, I meant actually into Q4 to Q1. Would you expect --

Q4 to Q1. Well, as we generally said in the past, and we still feel this way, the longer we move forward the more contracts we'll be entering gain-share model. And if we're successful with execution on our yield ramp engagements, we see no reason why our gain share should not continue to grow accordingly.

We get more specific information as we go through Q4, right. So then we'll provide a more specific dollar range at the end of Q4.

Okay, that's great. Thank you.

Bill Frerichs, D.A. Davidson --

Good afternoon. John, I have a question about the space that you occupy and some of the other competitive, or at least rhetorically competitive, moves that have been made in there, particularly Synopsys's buying IES, Cadence making a deal with ASML, and I noticed yesterday they threw Extraction into a DFM bucket when they reported their results. Menner (ph) announced extensions to Caliber that have a DFM argument behind them, and Magma of course is buying or bought Mojave, in addition to announcing a deal with you. And I was just wondering if you could try to elucidate what all this stuff is and what it has to do with your competitive picture, if anything.

Yeah, thank you, Bill. I guess I tried to address that in my script, but obviously I'm not -- I didn't do a great job.

There's a lot of buzz or hype associated with design for manufacturability, and now I see folks using words "holistic" and "yield ramp" in their press announcements as well. All of those are software tools to improve a design based on some model of the process. Our technology is really geared towards how do you model -- how do you know when you change something in a design how it's going to change the yield. So our -- what is unique about PDF is that characterization technology, the characterization vehicle test chip used to characterize a process and provide the information that you're going to need to make any of those choices and changes. So I think those -- you are going to see more and more folks coming out with tools that address making changes for designs to improve manufacturability. The net is cast wide. What is included in design for manufacturability today, as you alluded to, includes things as broad as extraction and LPC and actual yield. So it's hard for me to kind of directly compare to each of those products and technologies out there, because some of them have virtually nothing to do with us, or most of them have virtually nothing to do with us. But any time someone says they're going to change something to improve the actual yield, they need a model of how that yield -- how that change in the layout would affect the yield. And that's really what our technology does, provide that information, that model.

Okay, John, I'm sorry, I got off an airplane, didn't hear the first part of your script. So forgive me if you've already gone over this.

Well, I'm not sure that your question wasn't valid anyway.

Okay, great. That's helpful. Thanks very much.

Gus Richard, First Albany Corporation.

I just had a quick question on the tax rate going forward. How should we model this?

Well, you know, we are -- you know, tax rates are calendar-year events. As you wind down one year your forecasts become reality and your tax rate becomes a calculation. As you go into a new year you kind of reset the clock, you look at your forecast, you look at your tax credit considerations and then the other tax ramifications that might roll over into the new year. You know, I think going forward we're going to balance up towards the beginning nominal rate that we'll -- or I should say a beginning effective rate that we'll roll out across the new year, and we'll be somewhere under the standard effective tax rate as we benefit from some of the credits and things, but not dramatically better.

So as a percentage of pre-tax income -- should I model next year 30 percent? Is that a reasonable start?

Well, you know, I think it's going to be in the 30-something percent. I can't say that I want to point out an exact percent, but it will probably be higher than 30 percent going into the year, but more beneficial than the 40 percent as if we didn't have any research tax credits or any of those aspects to take into consideration.

And then for the December quarter?

You know, for the December quarter we're going to be seeing a little benefit of some of the credits that are going to come to fruition as we know where our year-end results are going to, and that's been built into the guidance that we provided.

Got it. Okay, all right, thank you.

Dennis Wassung with Adams, Harkness & Hill.

Thanks, guys. A couple more questions here, I guess one on the I guess the term here "cautious optimism" as you look into the Q1 guidance. I guess does that kind of imply that if the business you're tracking today does close as you expect that you're looking at a higher number than what your guidance is at this point?

Obviously the guidance is based on expected value of close, and because we percent complete, if contracts that we expect to close in Q4 and Q1 close earlier, than that does put us at least at the higher end of that range that we would provide. If they close later, that's at the lower end of the range. And obviously if they close higher or later, earlier or later from the window that we anticipate, then that would either -- that would cause you to set things up higher or set things down lower. Right now that range is based on some window we expect -- when we expect things to close.

And that range is based on the words that I used, "cautious optimism," that there is undercurrents in the industry that we have to take into consideration when we plan the expectation of when these contracts will close.

Well, I think if the macro environment or the environment is generally better, that would bring us to the higher end, and if it's generally less than we expect, it's toward the lower end.

Okay, fair enough. And looking at the gain-share contracts in the quarter here, it was the same number of contracting customers as it was last quarter. Was there any change in that, or were they the same the same contracts, the same customers?

Well, it's a similar configuration. The only thing that we pointed out was that, one, that was clearly expected to quick off in the quarter. The production schedules unfortunately shifted a little bit to the right, and it's been delayed until the fourth quarter.

Okay, so basically you're looking at fairly organic, if you will, increase to that gain-share number? That's pure volume increase?

Share yield, or you know in other words performance to target.

Certainly. Is the -- I guess when you look at that, the profile of those contracts, are those -- I guess are those mostly .18, .13 type of contracts in the mix at this point?

Mostly they're more leading edge than that, so I think 1.3, some 90, and probably still some residual 1.8s.

Okay, great. And the engagement you signed this quarter, was that a 90-nanometer deal?

Yes, it was.

So you're up to 9 total at this point?

I believe so.

Okay, great. And you also made a comment about one of your customers implementing pDfx in the quarter. I wasn't sure if that was a new quarter or was that sort of an existing customer?

Let me make that clear. So an existing yield ramp customer who we provided models for them to implement and design with pDfx during this quarter I think, and that's probably going to take out in Q4 and produce some time in '05, that we anticipate impacting that customer's gain-share number in the '05 timeframe.

So in an engagement like that, I guess, how does the pDfx piece impact your revenue? Is it just through the ability to get more gain share, or is there another sort of layer to the engagement?

Sometimes there's an incremental fixed fee, because we may need to help them use technology. But primarily what we're trying to do is get that to drive higher yields. We model what the yields are, and we can model roughly what that's going to do in terms of improvement above their targets and hence dollars to us. So I think we expect the bulk of -- we do generate some revenue from helping them use technology, or we will generate revenue from that, but we anticipate the majority of the revenue to come from the benefits they see in the yields.

All right, so pDfx is distributed to them through their EDA supplier, through the relationships that we have with that EDA supplier. The end result will be better yields, higher production, which will manifest itself in higher sharing of the value that our customer receives.

Is it fair to assume it's through the integration with the Magma tools?

Yes, it would be.

Okay, great. And I guess the last question, looking at sort of the new engagements you have in Q2, a couple of new customers there, any comments on sort of progress on those engagements? They were full yield-ramp engagements, I believe. Any commentary there?

You know, I said again during my script, with our new customers where they were able to see benefits from our test vehicles within the first production lot, and I think that's about all we can really say. So I think that is -- we design our whole system so the customers get results very quickly. I think that they're seeing that as we would expect.

Great, thank you.

Tim Fox, Deutsche Bank.

Steve, a couple questions for you, and then one for John. Can you just remind us what the operating margin targets are, if you have any for sort of medium and long term?

Well, let's see, I'm not sure how I could talk about that, but I would say from a strategic perspective, I mean, we're looking for 20 percent plus operating margins over the long term. I'd rather not comment at the moment what timeframe needs to be at what expense. That's a little more forward-looking a statement than I'd like to make. But you know clearly we expect to be delivering more and more profitability as we move forward.

And, related to that, you've announced a number of new partnerships with IP vendors, EDA and I guess some semi-cap partners. Are you anticipating any expenses, any extra expenses, on PDF's side of the gain to help these partnerships get through fruition? Is there any extra R&D effort needed on your part to do some characterization with Virage, for instance? Or is this mostly on the partners?

Well, I think John could speak to the technical capabilities relative to engineering effort. But what we have done is when we developed our project plans for the new year, part of our development efforts and the related budget includes people and technologies to help expedite those relationships. So I don't think it will be anything abnormal to our plans for the new year.

Okay, great. And, John, just for you, there's been some -- a little bit of mixed signals coming out of the environment around 90 nanometer. There were some comments from some backend equipment suppliers saying that 90 may not necessarily ramp as quickly, but EDA vendors are saying that they're seeing design activity move along fairly well at 90. What are you seeing from your perspective?

Yeah, so we still see it moving along pretty well. I think you know you probably saw inside Business News TI announced that they were moving over to 90 in the near term. That's something that I think is consistent with what we've been seeing across the broad spectrum of customers.

I think I don't see things particularly -- the yields are getting better and better at 90, so I suspect because the tool set is very similar between the 90-nanometer equip fab and a 130-nanometer equip fab. Once you're at yield parity, your superiority, it's very hard not to go there. So I suspect it will be pretty positive.

Okay, and lastly, if I may, you mentioned the consumer segment a little bit earlier. Have you seen any weakening there? We heard some of the IC device folks this morning pre-announcing some pretty bad news, and blaming it partly on the consumer sector. I know that's shipments, not design, but any weakness on the edges around the design world and the consumer world?

Yes, I think we use the term maybe -- that's a very big -- "consumer" is a very big catchall budget, so we would even consider the handset business as part of that, since it drives so much of nanometer volume, the consumer nanometer volume.

So far for the parts of the industry that we touch, would be the consumer segments that use the most advanced nodes -- 90, 130, potentially even 65 as you get out towards the end of '05 -- we still see a lot of activity, tape outs, beginning volumes, some more meaningful volumes. So at this point we don't see any frailty. I think a lot of times folks use consumer also to look at more trailing-edge technologies, you know, .18 and .15, lower, NDV systems for example. And we don't have much visibility to that part of the consumer space. So I don't know if we would be able to talk to it.

Gary Sendrow (ph) with J.P. Morgan.

Just looking back to the question about your first quarter guidance, just trying to understand you know if you look at it from the DSY and the DSY and the gain share, you know, if your gain share first quarter over fourth quarter is relatively flat, then your DSY business is only up a little bit. And if your gain share actually improves first quarter over fourth quarter then your DSY business is flat. So is there -- how should we think about that? I mean, is there a reason why that would be? Are there contracts running off for either side of the business in the first quarter?

I don't think there's much gain-share contracts that run off in that timeframe. We do always, you know, as you -- if your revenue base is growing, we do expect new DSY -- in other words, new fixed-fee portions of contracts to start. The timing on those, especially in the near year, which would be the situation in Q1, is always a little bit tough to predict, and Q1 at the end of the fiscal year in Japan as well, which also makes it a little bit more difficult to predict. So you know at this point, you know, we did say -- I think we did basically say due to timing we're not sure exactly where that -- when those are all going to start, and hence the guidance range is up over Q4, and we suspect that we're working real hard to make both components of the revenue go up, but at this point we're not sure exactly how it works out, mostly due to timing.

We've never really pointed out any cyclicality from quarter to quarter. However, the end of a calendar year does coincidentally have some contracts ending and new contracts are supposed to kick in, and the timing of those contracts would certainly have an impact on our first quarter. Simply, do we get a full 90 days of work? Is there any gaps? Is there any delays? You know, so that I think you've built into the dynamics of the first quarter forecast as well.

Okay, and just was it nine engagements at 90 nanometers?

Yeah, there have been nine ramps that we have done or we are involved with.

And can you refresh my memory on the customers that you've named at 90 nanometers?

You know, I think actually the only customer named at 90 nanometer is Chartered, maybe TSMC as well. I think everybody else, you can list customers often, but we have not made press announcements about whether it's a 90-nanometer ramp or not. So, you know, if you look at our K you will see customers that are over 10 percent. That does not -- you would need to guess whether those are customers at 90 nanometers or not.

Generally we have a policy that we don't identify customers and the specific work we're doing with them, unless they approve it. I think providing more color as to names and processes would be against our policy not to talk about that, unless it's approved.

Right. I was just trying to verify which were the customers that you named and which we're always guessing about.

And how many -- you know, you've got nine engagements at 90 nanometers, how many companies out there are out there today running at 90 nanometers?

That's a very good question, Gary. We constantly try to track that. There's still -- I think when we did the analysis, there's still double that number, depending on how broadly you classify it, that we will be at 90 or expect to be at 90 sometime over the next year or so, although we're not -- frankly, it's not a scientific number. So we were trying to figure out what is our market share at this point on a customer name basis or a fab basis, and we usually count by fab. And so it's at least -- it's at least 2x the number that we have so far, if not more.

Okay, and how much of that would be Intel?

Oh, a relatively small portion of it, I think. There's only a handful of fabs they would have, on the order of three or four.

Right, so your market share of X Intel goes up nicely?

So in that number I believe we didn't count them at that time. We -- (indiscernible) -- the logic at that point.

Okay, great.

A follow-up question from Bill Frerichs with D.A. Davidson.

John, forgive me if you covered this in your script, but could you contrast your deal with Chartered with the earlier deal with TSM, which had a very large upfront, because there was no follow-through really on wafers from their fabless semiconductor customers?

So I think both of them are tied to wafer volumes that are at capacity. In the case of the TSMC contract, there was no yield tied to it, so it was a license fee for use of the technology, but also a license fee period which was similar to the gain-share period proportional to the capacity of the fab at that note.

In the case of Chartered, you know, it's a traditional gain-share component contract where there is a yield ramp where we work closely with them to apply our technology for SOI and bulk. And in both process nodes there's a measurement of the yield performance, and thus establishes the wafer fee, and it's very much like the other contract in that there's the license fees, technology, for the production quarters. That's proportional then to the yield and the volume.

Yeah, to add a little financial flavor to that, the two contracts are not the same relative to accounting or revenue recognition, as we all painfully remember from when in the fourth quarter a couple of years ago we announced the contract as having revenue recognition over a multiple year extended period of time, which by the way we still continue to recognize revenue on in that fashion. The Chartered contract from a revenue recognition perspective is much more traditional PDF.

Okay, that presumes then that someone has signed a wafer level agreement with you?

Yes, that's correct.

At this time there are no more questions. I will now turn the call back over to Mr. Kibarian for closing remarks.

Thank you. PDF Solutions continues along its path of making our technology core to our customers' yield ramps. During the past quarter, with increased adoption at 90 nanometers, as well as our partnership with Virage Logic, we continue to gain traction in our core business. Thank you for taking time to participate in our third quarter 2004 conference call. Good-bye.

Ladies and gentlemen, this concludes the program. Thank you.