Good day, ladies and gentlemen and welcome to the PDF Solutions Incorporated conference call to discuss its financial results for the second fiscal quarter ended Monday, June 30, 2008. (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 to 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 demand for its solutions. PDF's actual results could differ materially. You should refer to the section entitled risk factors on pages 10 through 18 of PDF's annual report on Form 10-K for the fiscal year ending December 31, 2008, and similar disclosures and subsequent SEC filings. The forward-looking statements and risks stated in this conference call are based on information available to PDF today. PDF assumes no obligation to update them. Now, I'd like to introduce John Kibarian, PDF's President and Chief Executive Officer, and Keith Jones, PDF's Chief Financial Officer. Mr. Kibarian, please go ahead.

Thank you and welcome everyone. Today, I'm going to discuss the results for the second quarter of 2008 and discuss our outlook for the third quarter. For the second quarter of 2008, PDF Solutions is reporting total revenue of $21.1 million and non-GAAP net income of $0.03 per share. Gain share was $5.7 million. All of these results are in the ranges we provided in our guidance in April. Keith will talk more about these results and our guidance going forward in a few minutes. Our Q2 accomplishments continue to validate our growth strategy. With our expanded process life cycle solution, we are successfully driving more revenue per client by delivering expanded technology and securing longer gain share turns. As you may recall, in the second half of last year we signed a yield aware FDC contract, expanding our relationship with a long-standing yield ramp client by moving downstream into process control. Our yield to aware FDC solution addresses the problem of increasing process control costs which currently account for approximately 30% of wafer costs. In Q2, that client signed a multi million dollar extension for this solution. We are pleased with the acceptance that our yield aware FDC solution is achieving in the marketplace and the contracts signed over the last two years. Since we started offering yield aware FDC, two clients have engaged PDF for full deployment under contracts that can each generate revenues in excess of $10 million, a number of other clients are in the prove-out stage. We believe similar contract terms can be expected as these prove-outs transition to full contract. Not only have we increased the account penetration with our broader offerings, we are also increasing future revenue potential from gain share contracts. In Q2, we signed a 45-nanometer integrated yield ramp with a…

Thank you, John and good afternoon to everyone. Let me again state that this presentation and our press release issued earlier today include references 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 the reconciliations in the investor section of our website located at www.pdf.com. Revenue for the second quarter ended June 30, 2008, totaled $21.1 million, a decrease of 11% when compared to the second quarter of last year, but an increase of 4% from last quarter. These results were in the range we provided in our outlook press release on April 29. Compared to the second quarter of 2007, the decrease was the result of both a decrease in design to silicon yield solutions and gain share. Compared to last quarter, the increase was a result of a modest increase in both design to silicon yield solutions and gain share. Gain share was $5.7 million in the upper end of the range provided in April and represented our fifth consecutive quarter in which gain share exceeded $5 million. Design to silicon yield solutions revenue totaled $15.5 million for the second quarter, a decrease of 13% from the comparable period last year but an increase of 3% from last quarter. As John discussed, bookings were strong during the second quarter and represented our largest booking quarter since Q3 2006. We converted the L way from last quarter into a full 32-nanometer Logic Yield Ramp. We also closed three new engagements during the quarter. A 40-nanometer frost design kit, DSM Project, a 90-nanometer SOI project, both with existing customers, and a 45-nanometer logic ramp engagement with a new yield ramp customer, which began as a yield or FDC evaluation. The…

Thank you, Mr. Jones. (Operator Instructions). And our first question comes from the line of Matt Petkun from D. A. Davidson.

Hi. Good afternoon. It looks like a pretty descent bookings quarter. So, congratulations on that. I did want to ask, John or Keith, in the revenue breakout you’re no longer including license revenue. Are you bundling that into the solutions or really all contracts now signed as solutions and kind of related to that, how should we would be thinking about gross margin for that solutions line item going forward?

Well, Matt, we’ve been on that trend of combining more of our projects to combine all of our technology infrastructure, offering a solution, if you will. And as a discrete line item it was not meaningful and as we manage the business and we tend to sell in the future, we combined our software products with our services. So, on a go-forward basis, you should expect to see more of that, and that’s why we’re reporting it in that fashion. So overall, the margin projections are what we generally target as we’ve been doing that for some time now. It’s just a matter of changing the presentation of in the mid-60s or so for the current year and then our target margins as we’ve talked about as the company matures.

Okay. I guess I was thinking bouts the margins within what you’ve called solutions because you used to break out the cost of the licenses as obviously a separate, very small piece. So, it’s not that you aren’t ever signing your software licenses. It’s more that you’re just moving to report revenue like this, is that correct?

That’s correct. And when we go through our sales process, obviously as we introduce our products and our solutions, we see that we can add more value so it would be our choice, if you will, to sell them a solution offering versus a standalone software license. However, when the opportunity arises and that’s what the customer needs is, we do sell that on a standalone basis as a software solution.

Okay. And then my next question, last quarter you announced a pretty meaningful Data Power deal or at least it sounded meaningful on the call. Any update there, John?

Yes, that’s an example of where because of the services associated with that, it’s recognized more like a percent completion implementation, and a license revenue. We continue to deploy there and get positive feedback from them on the deployment. And as we get through this year we’ll get meaningfully through that first contract and look at how we can expand from there.

Okay. And then my last question, at least for now. Keith, you mentioned a 90-nanometer deal. I didn’t get what that was and I didn’t hear John talking about that. So, what was the 90-nanometer contract signed in the quarter?

Yes, I’ll explain that. We’ve done 90-nanometer contracts with the fab owners. By and large there are very few of those left, maybe one or two that we – or a handful or whatever. But now you have asset light manufacturers or effectively fabless at these nodes, and they want to ramp up a product or set of products in one of these factories that we’ve already characterized. We can provide them a mini ramp for them, which is fixed fees plus wafer fees. Our scribe line targeted to their products and services associated with that and some software that helps them bring up their products in that factory. And a lot of the benefit to them of course is we ramp that factory in 2004, 2003 or 2005 and have models and capability that make that go much more smoothly for them.

Is that something you’ve been actively marketing or are customers really coming to you for that?

We recognize that this would be an opportunity in the – Gees, when was that? I guess latter part of 2007, and we started talking to customers in Q1 of this past year and signed our first contract, basically in the beginning of Q2.

Okay. And then –

Something that would be out there and it was a matter of timing. We needed to have factories out there in the world at 90 and 65 that were starting to meaningfully talk to potential end customers.

Okay. And you said there will be at least some wafer based gain share fees?

We target that in this model and this contract has that.

Okay. Super. Thank you.

Your next question comes from the line of Mahesh Sanganeria from RBC Capital Markets.

Hi, guys. This is [Casey] going for Mahesh

Hi.

How are you doing, Casey.

Good. Couple of quick questions. You talk about two contracts of size larger than 10 million. Can you give a bit more color on those contracts, what is the kind of work, how many years and what is the prospect for most of these contracts in the future?

Casey that was on the yields aware FDC, there were other contracts that were multi-million dollars that have expected values in that range or greater. But the two that I called out in my script were the yield aware FDC contracts. Those all look a lot like our yield ramps in that there is deployment period which is typically four to six quarters or maybe as much as eight but typically four to six quarters, where we recognize revenue on a percent completion typically, and then there is a license phase where the license rate is targeted to the factory size or factory capacity. And runs for the period that the process control is running at the end of that period we turn off the process control. These are like our yield ramps in terms of the contract structure. What we’re doing is different. We’re deploying a system that is used to basically control or collect data off their tools like their etchers, deposition tools and evaluate our models. The models were built when they ran our test vehicles and basically identify wafers that need to be inspected or controlled or predict the actual metrology values so the customer can decide if that’s not a control wafer and as they run product on every product wafer, they have basically a measure of that wafer’s health as it goes through the factory. So, it’s like inspection but on every wafer, on every spot on the wafer instead of on a sample basis.

Okay. Can you talk about the approximate number of years?

The yield aware FDC contracts, Casey, they’re typically about a year of deployment, a year and a half of our teams working collaboratively with the customer and then the license period initially, the initial license periods are about two years and then there’s an ability for it to extend beyond that and that model, as John described it, it scales with their production. The two yield ramp engagements that we talked about, the 32-nanometer yield ramp engagement and the 45-nanometer yield ramp engagement, those are the typical terms that we traditionally work with our customers and actually the good thing and probably talk about that is that we’re actually pulls further back from R&D. Where the 32-nanometer is an eight quarter engagement and 45-nanometer is about a seven quarter, roughly, engagement.

Okay. Sounds good. You talk about fabless customers. Is that, should one consider that to be a growing category of customers I mean as opposed to simply focusing on the fab and foundries. Are you finding these fabless and the folks are sort of a new and probably growing source of future revenue?

Yes, Casey, this is the case. We do believe that and our strength, our value to them is the fact that we have characterization vehicles running in many of the factories that they’re considering running and hence can provide them characterization and things targeted to their specific products. It is a new opportunity for us and something that we are developing cautiously. We’re going to sell it to a few customers, demonstrate that we can add value and sell it to more. We’re not going to go higher and of course our expenses don’t show that, a huge sales force to go knock on a bunch of people’s doors, very targeted way to go after this opportunity. We believe it’s a significant and substantial opportunity. It increases our value to our process customers because the fact is we can help them be more successful with their customers. That gets them volume. Helps us with gain share on our base contract as well as creating an additional revenue source for us.

Okay. Can you give us a sense of your guesstimate for wafer starts in the second half?

Yes, so we guided Q3 as far as our gain share for us roughly flat. When you peel the onion on that, we see some factories increasing their volumes and some factories decreasing their volumes. Fundamentally I think the end customers, in other words the fabless chip companies and the product groups, tend to favor the larger factories, so you would think that a smaller factory would be easier to fill than a larger factory but if you look at the utilizations that we see in the customer base, the bigger the factory, the higher the utilization and the smaller the factory, the more they struggle with utilization. It could be indicative of their intrinsic cost disadvantages. For the remainder of this year, we do think we’re going to see that same scenario play out where either because of the end market that the factory is targeted at or its competitive position, it will do better or worse. This is very consistent when you look at UMC, SIMC and chartered announcements over the past week, you see kind of a mixed up and down or flat capacity utilization across the foundry base. Now, we’re not saying those or any of those are our customers but that kind of gives you an indicator of the overall industry. We’re suspect Q3 is going to be like that. We’re getting our first looks at Q4. We don’t think it’s materially different than Q3 yet. We will evaluate that as we go through Q3. As far as the memory market is concerned, what we saw in Q2 was, I think, a rather disappointing. Of course the flash manufacturers went through a relatively steep price decline and in Taiwan we see a smaller number of folks making investments than we originally forecast in the second half of this year. We think that’s partly due to the difficulties in the licensees transferring nodes that would be profitable to them so all the 50X nanometer nodes we think the transfer into those factories is going to be difficult. That will create an opportunity for us as that market recovers and the 70X nodes are not economically competitive to run at this point so we’ll see very little investment in that part of the market at this point.

Okay. And lastly, how should we think about share count into Q3?

Share count in Q3 should be relatively stable. I think it’s going to be as it was in Q2, it’s going to be a hair below 27 million shares, 28 million shares rather. I think we’re projecting about 27 million, 800 shares of our share count as we wrap up Q3.

Okay. Thanks a lot, guys.

Your next question comes from the line of Gus Richard from Piper Jaffray.

Hi, Gus.

Hey. Couple of quick questions on the yield aware. How many customers do you have in that product category at this point?

Three customers.

Three customers. And what is it a NAND, a DRAM and a logic? What flavor of customer do you have at this point?

There are definitely, actually, I think it’s more than three but there is definitely one memory customer in there and a handful of logic customers.

Okay. Okay. Got it. And then in talking to the foundry folks, there seems to be quite a bit of difficulty getting to 40-nanometers at some of the fabs, particularly in terms of power dissipation, and you’re seeing the same thing. Is that creating or opening some doors for you that haven’t been always opened in the past?

Yes. It’s definitely creating opportunities for us. I think you’re, yes, you’re right. The issue for 40 and will be the issue going forward, the power consumption is tricky. The SRAM is particularly tricky. The very aggressive bit cell sizes are hard to build, and on the logics side, the standby power is getting to be a bigger and bigger issue. That’s why we targeted BRICs last year when we bought BRICs, because the prove out that we demonstrate to one customer in Q1 is we could reduce your standby power by 20% with this type of technology, over 20%, and reduce you chip variation significantly because you’ve got a very simple or reduced set of layout patterns. It's also providing opportunities for us on the yield aware FDC. One of those two greater than $10 million contracts, that was really a customer who was pretty forward thinking saying we're going to have to control our parametrics tightly. Our production control gain is going from being a defect only gain to really being a parametric control gain and that's really why they bought. They were very forward thinking. We see this is going to make our process control solution more valuable to account. And on the yield ramp business, yes, we see, again, this has been creating interest in the customer base. In particular, our scribe line solutions because every customer uses the process differently. So the factory can often demonstrate very good power consumption on a test vehicle, a test vehicle SRAM, and then they put the customer's product in and they see things are really different. That's why these little 90-nanometer -- it's no so little, but fabless product ramp is really going to be kind of the wave of the future as we're going to do things targeted to each of these customers.

Got it. And then if I'm not -- just in some of the people you've worked with at the 40X node in the foundry space seem to be doing a little bit better than some of the folks you haven't been working with historically. Have I got this right?

Yes, yes. We are very happy when we see clients do really well and we work hard for that and we do think it highlights their enlightenment, maybe partly due to us and partly due to other enlightened choices they've made but for sure we're very proud of that.

Okay. And I would suspect that that's driving some business to your customers at that node.

Exactly.

Okay. Alright. I've asked enough questions. Thanks so much. I appreciate the help.

Sure Gus.

And you have a follow-up question from Matt Petkun with D.A. Davidson.

Hi. John, I was wondering if you could use, now that you have this 32-nanometer deal, the LOA from last quarter signed, if you use that as a bit of a test case for 32-nanometer engagements, assuming that you worked with this customer at 45 which I think you did, and 65, can you help us understand how your opportunity with each individual customer is changing from node to node? It's really tough from where I sit at least to see incremental penetration at any given customer both in terms of the number of devices that you're working with or the scale of individual IYR contracts, so if you could maybe address that.

Sure. So this is the fourth node with that customer. And in terms of the contract, it's longer in length. It gives us more potential, especially on the gain share period is much longer so it gives us more potential to collect on the production side. One of the risks in our gain share is when -- how much volume do they get and whether they get enough volume that we can get back a good return on the contract. Our contract profit really comes greatly from how much they build. Every time we lengthen out the contract or adjust the terms of the gain share generally improves our opportunity with those accounts. Of course they still need to win business and that is in their hands. The business opportunity or the increase with that customer or any other comes in three areas. And we're starting to see this with some of our oldest customers, first, as we sell, you get into the second, third and fourth contract, you get a layering effect on the contract which improves the total revenue you generate out of the customer. Second, we've been going back to those customers and showing them our BRICs technology and because they have our test vehicles, we can demonstrate to them what would be the advantage if their designers were on a BRICs like platform in terms of things, as Gus brought up, like power consumption. That gives us an additional upsell. And, thirdly, we've been going back in and showing them our production control solution and saying look, we can deploy this in your factory, capture -- we may be on one or two nodes in that factory. That factory may be running three or four nodes. We can capture those other nodes and importantly what this also does for us is continue the length of time that we're getting paid on some of the most old nodes. So we continue to generate revenue at 65 and 90 and 130-nanometer nodes because they're all being run in that factory. And that mitigates our risk in terms of what node ends up hitting in terms of that factory volume.

Okay. And then I suspect if we know who this customer is or if we guessed, there may be other partner from an R&D perspective fabs coming up and I think one of those specific folks actually talked up their expansion plans in logic for next year 2009 and ramped up their CapEx last quarter or last week. I'm wondering when you think the opportunity to win some of those types of partners might be.

Yes, that's a very good point, Matt. The -- our oldest customer was Toshiba and they went on a path that fundamentally we see the other customers go down. They drive it earlier, earlier, and earlier and their technology development as they realize they could use our systems to really make the technology itself more manufacturable. In the case of Toshiba, that really happened as they went from the 130-nanometer node down to the 65-nanometer node and they continue with us beyond that. That's happening in this case just now. You can look at the case of Toshiba, when we would come in later in the life they would already have their own test vehicles that they did development on and then they would eventually turn around and use ours as they went and transferred the manufacturing. In this case, by selling earlier, we really get our vehicles inserted earlier in the development process. It gives us the ability -- then use our basically for significant portion of development, one of their licensees and, unlike Toshiba, some of the other companies are much more aggressive licensees. They have quite a few licensees. When their licensees look to inbound that technology, our vehicle set is now the measuring stick and we find that to be very powerful. We've been exploiting that at the 45-nanometer node where we allowed alliance partners to be able to compare and understand how they manufacture each aspect of the process relative to each other. This will give us, I think, a bigger opportunity to go and get a higher percentage of those licensees.

Okay. And then you mentioned the fabrics technology a lot on this call which I find encouraging. Do you really see 32-nanometer SRAM as the catalyst or the primary catalyst for fabrics? How many deals have incorporated fabrics IP thus far?

We have two classes who have incorporated fabrics IP so far and a number of others in early eval stage. It's not the SRAM per se. The fundamental issue is going to be when we acquired fabrics, we basically said to ourselves there will -- 193, high NA stepper will be the stepper of choice at 32 and at 22 and I think potentially even a node after that, whether it's 16 or not. That depends a bit on timing. Hence, you are going to have a challenge on doing design as the way you used to do it. Doing design with the set of design rules that are used to build standard cells and standard logic will become more and more costly. It will require double exposure. It will require a lot more power consumption and a lot more within chip variability that will drive up the complexity of your design flow. You're going to see more verification on the parametrics side and more verification on the printability side. If you go to a fabrics like layout, what we're able to demonstrate to the customer is your power consumption problems go away, your thin chip variability problems go away so you don't need to do as much verification as you thought you need to do and your lithography challenges are greatly ameliorated so you can take a layout one node and optically shrink it to the next node and it clears your ORC violations, it's ORC free.

Okay.

So that will I think greatly change the cost profile on 32 and 22, which will be the limiter on how many folks and what products go to those nodes.

Okay.

It's not that we can't do it's just how expensive is it to do it. It really comes down to how simple can you make lithography and what we're able to demonstrate with BRICs is you can make it very simple.

Great. Thanks so much.

Thank you.

Your next question comes from [Gary Shinaro] with JP Morgan.

Hi, John. Hi, Keith.

Hey, Gary.

Did you say how many contracts, Lucent’s contracts rolled off in the quarter and also how many you expect to roll off in the third quarter?

No, we didn't mention it during the call. I think we had maybe one in or two. One was rolled off, replaced and then a smaller contract and in terms of the falloff into the future quarters, there is nothing significant that we initially foresee.

Okay. And when John, when you were talking about the Data Power contract and that being percentage of completion, so are you saying that you haven't recognized much revenue on that in these two quarters and, if so, when do you think you should start seeing that?

Okay. And when John, when you were talking about the Data Power contract and that being percentage of completion, so are you saying that you haven't recognized much revenue on that in these two quarters and, if so, when do you think you should start seeing that?

So I will answer that, Gary. For the Data Power contract that we mentioned last quarter, because of the complexity, the amount of service that was involved, the customization of the software, we recognize that in a very similar fashion that we recognize a yield ramp project. So it is a matter of what resources we commit to during that period versus our total estimated cost to complete. We initially anticipated that contract being fairly evenly staffed, if you will, and the work efforts to commence through the end of the year. So we have recognized revenue both in smaller amounts in Q1 and a little bit larger amount in Q2 and then there will be some revenue contribution in Q3 and Q4.

And when you say some revenue contribution, meaning it will be – it is building from first quarter to second quarter to third quarter to fourth quarter?

And when you say some revenue contribution, meaning it will be – it is building from first quarter to second quarter to third quarter to fourth quarter?

Not necessarily, no. We have kind of differences in staffing levels, but my point I was trying to make is that there is revenue coming from that contract over in essence of multiple four-quarter period of time versus typically when we had sold a typical Data Power software track and we recognized all that revenue in one quarter. So there is ongoing revenue contribution.

Okay. Got it. Thank you. And you had obviously a nice quarter in terms of new engagements signed. I think I missed part of it. Could you just go over what the engagements were and did those -- can you just talk about your pipeline like usually you have good visibility on the pipeline. So the good number of engagements, did that deplete the pipeline or does the pipeline remain big?

Okay. Got it. Thank you. And you had obviously a nice quarter in terms of new engagements signed. I think I missed part of it. Could you just go over what the engagements were and did those -- can you just talk about your pipeline like usually you have good visibility on the pipeline. So the good number of engagements, did that deplete the pipeline or does the pipeline remain big?

I will take the pipeline part and then Keith can answer the engagements. So, when we -- our Q1, we basically said we lost business, we think it just took longer to close. That's why we had, as Keith said in his script, a very large bookings quarter in Q2. We believe we will have a good bookings quarter in Q3. We don't think it will be as good because we believe some of the Q2 bookings was a catch-up from Q1. And that kind of speaks to the fact that we didn't think at that time that we lost business, it seems in hindsight that we haven't. Now in terms of what we file, I'll turn it back over to Keith.

Okay, and just to be clear, because I know it might be a little difficult to catch all the deals, especially when there is a good number of deals in any particular quarter, but just kind of starting off, cleaning up from where we had left off last quarter, we had an LOA for 32-nanometer logic engagement for a integrated yield ramp that we finally completed in the quarter. In addition to that, we signed three new engagements. We had a 40-nanometer DFM project, we had a 90-nanometer SOI project, and then we also had an additional 45-nanometer logic integrated yield ramp. So those were the three new ones that we signed in the quarter. But in addition to those items I just mentioned, we had an existing yield aware FDC project, where we in essence, more than doubled the bookings by extending that contract, adding more services, adding more value to the customer that we completed in the quarter as well. So it was a fairly meaningful and significant bookings quarter all of which we had in our forecast and we executed upon in the quarter.

Okay. Great. So the LOA, the 32-nanometer LOA that went to contract was on top of the three new?

Okay. Great. So the LOA, the 32-nanometer LOA that went to contract was on top of the three new?

Right.

Okay.

Okay.

Q1 LOA that converts to a contract.

And just last question, regarding the share count being flat, I would have thought that would be forecast to go down with the continued buyback.

And just last question, regarding the share count being flat, I would have thought that would be forecast to go down with the continued buyback.

So last quarter we did buy back 283,000 shares, and we did see a little bit of a blip going down. And as we go -- we are building our cash, we are doing things, but in the forecast we are forecasting it to have some of it going down. A year ago we were in excess of 28 million shares.

Yes, I think every quarter, the last three or four quarters we have bought back on the order of 200,000 to 500,000 shares a quarter. We don't make claims about what we will or won't do in any given quarter. But, you can -- we factored some things into our model and you can certainly do the same on yours.

And you have a follow-up question from Matt Petkun with D.A. Davidson.

More than my fair share. So, first, you called that 90-nanometer deal an SOI deal like silicon on insulator or did I mishear you?

I just refer to it as 90-nanometer logic. Keith provided a little bit more information and said it was a SOI logic process.

Okay. Fair enough. That's good information. And then Keith, did you guys generate cash from operations in Q2 and what's your six months cash generation if there is any thus far?

We did generate cash from operations this quarter. It was about $2 million that we generated in the quarter. In Q1, it was slightly to the negative in terms of cash from operations but for on a year-to-date basis we are positive cash from operations and for the year we continue to expand on that trend.

Okay. And then finally, because I haven't heard anything mentioned about this, Keith, can you comment a little bit more about [Joy Leo] joining and kind of what strategic or management related changes are being made organizationally?

Sure. Joy is a wonderful addition to the team. As our company grows, becomes very more complex, and as we see as we try to venture into new products, new solutions, it is very apparent that we need to free more of John's time to be customer facing. In addition to that, in this new environment with the complexities that we have from compliance regulations, SEC reporting, general financial management of the company, it was a very good and reasonable fit to add an additional support in that regard. And, as you know, Joy has a wonderful skill set, a great track record and in addition as we went through and took a look at the business, from a cost perspective and we are very cost conscious company, a substantial portion of our cost expenses do come from the HR or headcount side. So we recently had a turnover in our VP of HR and it was kind of an opportune time to kind of rethink and realign how we are going to manage the business. We were very lucky to have joy available at that time and she is truly a breath of fresh air and we are all very excited working with her. I am very excited partnering with her, learning from her, and her adding to the team.

Okay. Great. Thanks, Keith.

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

Thank you for joining our conference call. Good-bye.

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