Quicklogic Corp (QUIK) 2006 Q2 法說會逐字稿

Welcome to the QuickLogic Corporation second quarter fiscal 2006 conference call. Today’s call is being recorded and will available for playback beginning one hour after the completion of the call. [Operator Instructions].

At this time for opening remarks and introductions, I would like to turn the call over to Mr. E. Thomas Hart, Chairman, President, and CEO for QuickLogic. Sir, please go ahead.

Good afternoon, ladies and gentlemen. Carl Mills, our CFO, and I are here today and welcome you all to our Q2 fiscal 2000 conference call. We appreciate you taking time to join us and hear about QuickLogic’s revenue results and our continuing progress in providing the world’s lowest-power programmable logic solutions for high-volume applications.

In today’s press release, we noted we’re conducting an internal review of our stock option practices and related accounting. Our Audit Committee asked management to initiate this review. While we’re not alone in this endeavor, it doesn’t feel good to postpone providing you with our full financials. We are focused on completing this review and getting back on track with our financial reporting. This review could impact our ability to file our 10Q in a timely manner. As we get further into the review, we may announce a delay in this filing. In the meantime, our officers and employees are not commenting on this review.

Carl with take you through our Q2 2006 revenue, and I’ll share my perspective on our business. Finally, Carl will detail our revenue guidance for Q3, and then we’ll take your questions. Carl?

Thank you, Tom. Before we get started, I’d like to read a short Safe Harbor Statement. During this call, we will make statements that are forward looking. These forward-looking statements involve risks and uncertainties, including but not limited to stated expectations relating to revenue growth from our new products, statements pertaining to our ability to convert new design opportunities into customer activity, market acceptance of our customers’ products, the impact of regional conflicts on our suppliers and customers, the effect of the options review on our financial statements, and the possibility of a delay in the filing of our Form 10Q. QuickLogic’s future results could differ materially from such forward-looking statements. We refer you to the risk factors in our annual report on Form 10K, quarterly reports on Form 10Q, and prior press releases for a description of these and other risks that could cause actual results to differ materially from our forward-looking statements. QuickLogic assumes no obligation to update any such forward-looking statements.

For your information, this conference call is open to all and is being Webcast live. It can be accessed from the Investor Relations area of the QuickLogic Website located at www.QuickLogic.com/Investors.

New product revenue in the second quarter of 2006, which includes our Eclipse II™, QuickPCI II® , PolarPro™, and QuickMIPS products, more than doubled, to $2.3 million from $1.1 million in the first quarter and represented 25% of total Q2 revenue. We received strong customer demand during the quarter, and our Operations Group did an excellent job of delivering on the opportunity. On our April call, we said that we expected new products to contribute 16 to 20% of revenue. Taking the midpoint of our total revenue guidance at that time, this would mean new products would contribute up to $2 million of revenue. Our actual new product revenue was $2.3 million, significantly better than that guidance.

Second quarter total revenue of $9.2 million, however, was lower than our April guidance of between $9.7 million and $10.3 million. As noted in our July 12 press release, we had sufficient demand to make our original guidance, but we were unable to ship two large pASIC1orders. In one case, the customer changed its requirements less than two weeks from the end of the quarter, requesting that we program the devices and that the customer conduct source inspection. We simply couldn’t meet this customer’s new requirements before the end of the quarter.

In the second instance, wafer deliveries from our supplier were late. Both of these orders are now forecast to ship in the third quarter. We had guided that revenue from pASIC1 and pASIC2 products would be between 12 and 15% of $10 million, or $1.2 to $1.6 million. Actual revenue from p1 and p2 was $1.1 million. Additionally, orders from some of our other products, such as QuickRAM and p3, were lower than we expected during the quarter. Overall, higher new product revenue was insufficient to make up for the delayed pASIC1 shipments, and we could not make up the shortfall by shipment of other products.

Also of note, we have two significant customers in the second quarter. A domestic OEM, purchasing primarily pASIC3 devices, accounted for 12% of revenue. A European customer, purchasing [inaudible] products through a contract manufacturer, accounted for 19% of revenue.

Let’s just take a moment to compare our results to the same quarter a year ago. Our revenue declined $3.5 million year over year. Revenue from our pASIC1 and pASIC2 products, which are completing an end-of-life program, declined by $3.9 million. Revenue from our other products declined by $1.7 million compared to one year ago due to lower sales to a large customer in China, and a large customer in Japan. These declines were partially offset by the $2.1 million increase in revenue from our new products.

Now, for the other good news, our second quarter ending cash position of $30.2 million is flat compared to the last quarter and is better than our April guidance that we would use $2 to $3 million of cash in the quarter. There are two significant reasons for this improvement. First, we moved spending associated with a new manufacturing agreement into Q3, and secondly, cash received upon the exercise of stock options was greater than we expected in the quarter. Our debt-free cash position of $26.7 million dollars increased by half a million dollars during Q2. And this was our sixth consecutive quarter of increasing debt-free cash. We also renewed our Silicon Valley Bank line of credit during the quarter. Because of our strong balance sheet, we reduced our revolving line of credit to $5 million from $8 million. In summary, although we are disappointed with our total revenue for the quarter, we were pleased to more than double our new product revenue sequentially and maintain a strong cash position.

Now I’d like to turn the call back to Tom.

Okay, thank you, Carl. Well, what can I say? From a top line, as in revenue perspective, Q2 was certainly disappointing to myself and the dedicated team here at QuickLogic. This is our twenty-eighth conference call, and we’ve only pre-announced twice since we went public in October of ’99. This time we missed our top-line guidance by less than 5%. The truly frustrating part of this miss is we had the business booked, but pASICI shipments let us down. The fab was late delivering wafers, and a big customer changed his requirements late in the quarter.

So this is the good news for the quarter: new products accounted for almost 25% of total revenue, up from 12% in Q1. This means new product revenue more than doubled sequentially on a dollar basis. Also of significant note is the fact that advanced ESP revenue was up 54% sequentially, and 53% year over year. So, more importantly, you might ask, “Where is this new product revenue coming from, and how much more can we see of it this year?”

Those are easy questions to ask, not necessarily as easy to answer them, but let me give it a try. First, in terms of whom we’re shipping to, last quarter we shipped new products, that’s Eclipse II™, QuickPCI II®, QuickMIPS, and PolarPro™, to over 40 different customers worldwide. These include our traditional instrumentation and test customers, like Honeywell, Guidant, Abbott Labs, and Yokogawa, plus our traditional military customers like General Dynamics and Rockwell Collins. Our gaming customers, those that prize our bulletproof intellectual property security, included Konami in Japan, and a long-term customer in Italy. And for our new market consisting of professional consumer, but we call them prosumer customers, we shipped both production and pre-production orders to two of the Tier I ODMs in AsiaPac. Our single largest customer for our new product remains an OEM customer in Europe who’s building a 3G cellular data card family.

Now, for the tough part of the question, “How much more of these products can we expect to ship this year?” Well, the volume ODMs, the ones to whom we’re supplying a solution and not just an FPGA, are new customers for us. Since we’ve not done business with these customers, we don’t have a good feel for the way they do business, especially when it comes to their demand forecasting accuracy. Actually, this makes our revenue forecasting a distinct challenge.

We have been told by our customers that will be on store shelves by this Christmas in (1) a high-end GPS system, (2) a most exciting smart phone. In fact, when I visited the ODM in June, they told me I’d definitely want one of these, being the high-tech guy that I am, of course. And finally, an ultra-mobile personal computer. These products aren’t pipe dreams. But as we’ve seen, demand is very tough to call, from both our customers and therefore for us. We do have the good fortune to have wafer capacity and recently secured guaranteed assembly capacity. We also have sufficient test and programming capacity in place that could allow us to react very quickly to a significant increase in demand. I, for one, look forward to having that challenge.

In Q1, we achieved full production status for the first two members of PolarPro™ product family, the 75,000 and 100,000 gate devices. In Q2, we began sampling the 300,000 gate product, and this device is now slated to be in full production this quarter. As you may remember, PolarPro™ was voted by the readers of EDM Magazine to be one of the hottest products introduced in 2005. And just this week, we got the Cool Beans Award—that’s for real, by the way—Cool Beans Award at the Design Automation Conference in San Francisco. Cool beans, indeed.

Let me quote Max Maxfield, the editor of SOC Central, who made the award. “The folks at QuickLogic leapt onto center stage with their new PolarPro™ family of FPGAs that draw less than 10 microamps when in the inactive mode.” You may remember that the primary innovations of PolarPro™ are aimed at continuing our customer heightened awareness of having the lowest power FPGAs in the industry. We truly began with our Eclipse II™ family. Eclipse II™ was the lowest power FPGA in the industry, and PolarPro™ draws half the current that Eclipse II™ uses. We shipped the first production orders for PolarPro™ during Q2. As I mentioned last fall, this includes a customer that builds portable, battery-powered medical monitoring products, where battery life is, of course, most important.

Now let’s shift gears here for a minute and talk about the Intel Xscale embedded processor family and the recent announcement that Marvell is buying that product line from Intel. While we’ve certainly enjoyed and gotten significant benefits from our Intel collaboration, we look forward to engaging with the very aggressive Marvell team. I was recently in Taiwan and Mainland China, visiting both ODM and OEM customers that were using Xscale processors. None of them voiced concern regarding the potential of Intel selling the product line. And if you heard, or go look at the transcript of Marvell’s conference call announcing the purchase, I think you will agree Marvell has done a good job, a very good job, actually, and is making for a smooth transition. This includes securing the key members of the Intel Xscale team to join Marvell.

Investors have voiced two major areas of concern for our business as it relates to the Marvell acquisition of Xscale. One is that this change of ownership might cause a delay in growing our short-term revenue, specifically revenue over the next several quarters. At this point, I think this is unlikely to cause us problems. Our customers are committed to Xscale, and Intel has announced it will continue to provide wafer capacity for Xscale processors to Marvell for the next two years.

The second area of concern we’ve heard from investors centers around Marvell being a very quick-moving company that might want to integrate the benefits we provide into their own silicon. I would suggest to you that this phenomenon is fundamental to the very nature of the semiconductor industry. Products made today requiring five devices sooner go with three chips, and then just one, all over several generations of product. This process is what has made the semiconductor industry the only industry in man’s economic history where the cost per function has continually declined over the life of our industry. Internet LANs are a perfect example of this reality.

Our answer here is to make sure we continue to innovate with solutions that our customers need. The key advantage we have is that our time to market, with key solutions that are needed and economically viable for both our customers and us, is much shorter than our competition. Why is this? We use both field programmable and hard-wired logic in our embedded standard product device architecture. And please don’t forget that we’re providing complete solutions here, including silicon and software. And by all means, don’t forget that we have the lowest-power field programmable logic in the marketplace, bar none, which is an obvious must-have for companies making battery-powered products.

Well, let’s wrap up today on our prepared remarks. Yes, we are still navigating our way through the nasty bit, until revenue from our new products can completely replace the revenue from our end-of-life and some older products that are going away. But in our minds, it is still very much not a question of if this will happen, but only a question of when it will happen. Carl, back to you.

Thank you, Tom. Now let’s turn to our revenue guidance for the third quarter and some comments on our business. As part of this guidance, we expect new product revenue, Eclipse II™, QuickPCI II, PolarPro™, and QuickMIPS, to be flat to up low single digits. New orders for new products will be interesting. Consumer applications began in late Q2, and these customers are new to us. Because this business is new to us, the timely ramp of revenue from our new products is difficult to forecast. For example, we had expected to book new product orders for a handful of consumer applications in the second quarter, and to ship these orders in the second and third quarter. And we did receive production orders in the second quarter for a GPS application and pre-production orders for a smart phone. We believe revenue from the largest customer for our new products, a European customer, will temporarily decline in Q3, and that this decline will mask growth from our other customers.

Given all of these factors, we strongly believes in the growth of our lowest-power business, and just our questions about the when, as Tom explained. We expect our pASIC1 and pASIC2 products to contribute $600,000 to $800,000 of revenue in the third quarter, based on the shipment of the two delinquencies mentioned earlier. This represents a $300,000 to $500,000 sequential decline attributable to the end of life of these products.

I want to take a moment to again confirm that we expect the foundry to take the equipment used to manufacture these devices out of service this quarter. The p1/p2 business is truly coming to and end for us, and any significant upside to our current outlook is limited by the product we currently have on hand or on order.

Finally, looking to some of other products such as QuickRAM, were softer than we expected in Q2. While bookings for our other products have increased so far this quarter, Q3 is typically a tough-looking quarter for this business owing to European vacations. Overall, we expect that our total third quarter revenue will be between $8.3 and $9 million.

In light of our outlook for third quarter revenue, we currently do not believe we will see sufficient demand to maintain our guidance for revenue growth in 2006 over 2005. Now I would like to turn the call over to Tom for his closing comments.

Okay, thank you, Carl. We’re very pleased to have a new Director join our Board. Nicholas Aretakis, the Vice President of Worldwide Sales at Advanced Analogic Technologies, brings a strong background in semiconductor sales and marketing expertise to our Board, which, when coupled with his extensive AsiaPac customer knowledge, will certainly bring a significant addition to our team. Welcome aboard, Nicholas.

Again this quarter, I would also like to take this opportunity to thank our dedicated employees, whose talent, skills, intellect, and tireless efforts make our accelerating progress possible. Our teams all around the world are performing at unprecedented levels. For example, PolarPro™, our newest award-winning low-power FPGA product family, lowest-power FPGA product in the industry, bar none, came to market in half the time that the previous Eclipse II™ took. That’s time from tape-out to full production, a truly superb effort with clearly outstanding results. Thanks to all of you.

Now, for scheduling purposes, we’re presenting at the SRA Tech Conference on August 7 in San Francisco, at the Cannacord Adams Summer Seminar in Boston on August 8 to 10, and at the A.G. Edwards Growth Conference in New York on September 20. Our Q3 earnings conference call is scheduled for Wednesday, October 25, at 2:30 p.m. Pacific Time. Details of upcoming events can be found on our Website at www.QuickLogic.com.

Okay, now let’s open up the call for questions. Sherry, please?

++ q-and-a

Thank you, Mr. Hart. The question-and-answer session will be conducted electronically today. [Operator instructions]. We’ll take our first question from John Lau from Jefferies.

Yes, Tom. I wanted to follow up with you with regards to the new product. What kind of business, I mean, probably the toughest part is to really see what is happening. What are you seeing in associated design activity and the potentials as you see going through the September quarter, and what do you feel is ahead of us for the December quarter? Thank you.

Okay. Well, design activity depends on where you are in the design funnel at this point. The design cycles that we’ve seen for these products are on the order of nine months. So there’s not a lot of, if we’re just starting at the front end of designs at this point, there’s not much chance that you’re going to see much revenue between now and the end of the year. So PolarPro™, we’ve talked about in the past, we don’t expect to see significant revenue. There’s potential for it in the fourth quarter, but we really haven’t expected to see significant revenue until Q1 of next year. So this year will be carried by Eclipse II™, QuickPCI II, and QuickMIPS II, and the design activity there in the past has been solid, and we believe we have a shot at making good on our previous guidance. Looks like it’s a stretch at this point.

So the basic problem is design activities that really hadn’t started about nine months ago. Tom, how do you feel with regards to the potential that looking forward it’s not just on the December quarter because of the low visibility, but what do you feel the potential will be in the December quarter, because we really need to get through this period, especially for the September quarter, to see what’s beyond that.

Boy, I agree with you there, John. You know, the challenge we have is the one I commented on, and that is we don’t have a lot of experience with these guys. So we don’t really know—let me give you the example of the European guy. I think I’ve done this before, but I’ll share it with you again to show you the kind of challenges that you can have. I visited the customer in April, this was 2005, they told me they would need 30,000 units between then and the end of the year. My sales VP visited them in June, and they said they’d need 60,000 between then and the end of the year. We shipped them almost 400,000 last year. We’ve shipped them almost a million to date since the program began, which far exceeds anything that they had projected. So that’s one example. I would hope that we would be troubled with those kind of examples on a forward-going basis, but I can’t bet on it. I can tell you that the projects that we’re designed into are all, we’re looking at 150,000 to 500,000 units per year, and there’s lots of projects that we’ve been working on. So we feel good about where this thing is going. The question is really, as we pointed out before, the question is when, not if.

Okay. Thank you.

[Operator instructions]. Our next question comes from Charlie Glavin with Needham and Company.

Thanks. Carl, I apologize, but if you could just go over the guidance again, or you said that new product would be single digit. In terms of the older products, given the carryover, will all this carry over, and could you give some explicit guidance as far as that sector?

Sure. Overall, our revenue guidance was $8.3 million, Charlie. And our new products, which were $2.3 million last quarter, we said we’d be flat at low single digits. And what we’ve really got going on there are two dynamics. We expect our large European customer to temporarily have lower revenue in Q3 and to rebound in Q4. We think that that decline is going to mask growth from our other customers and put us net low single-digit range. That’s what’s going on there. Now, in terms of p1 and p2, we just came off of a $1.1 million quarter, and we said that we’d be down $300,000 to $500,000 sequentially. Actually, what’s happening with p1/p2 is we’re carrying some backlog that we couldn’t ship last quarter into Q3. We expect to get that out the door this quarter, no problem. We may get a little bit more business, but we’re not counting on it. Okay? And, finally, the other part of our business was softer than we wanted from a bookings perspective last quarter, and that caused us to have a little bit dampened outlook for Q3 relative to last quarter. And that’s basically the dynamics that we’ve got going on.

Yes, to that last point there with more of the [inaudible] and during your pre-announcement you indicated that softness. Could you go into a little more detail? What exactly are you talking about? Certainly the lumpiness as far as new products, particularly with some of the channels that we’ve seen that was Broadcom and other companies, also reporting, but in terms of the softness for the embedded [inaudible], more stable type of ramp. If you could go into a little more detail about what you’re seeing with that particular group of softness.

Well, we think what happened is we had a temporary reduction in bookings last quarter. By the same token, Q3 is typically a tough quarter for us in terms of bookings, so we’re reluctant to be bullish on how hard that booking rate can rebound this quarter. And that’s why we’ve got the cautious outlook.

And, in effect, Tom, you were using this analogy of several orders or indication within the 150 to 500 range, and a very unusual move, Bill Ruehle of Broadcom actually gave an Eclipse dollar amount or indication for the fourth quarter. When you were giving those numbers, were you also seeing [inaudible] three customers or three very large volume orders. What is your confidence level that those three will come in, and are they of that magnitude of several hundred thousand units?

Well, we talked about the 150,000 to 500,000, Charlie, that’s a per-year kind of level. And it’s per model. So, for instance, our datacom customer, we’re in multiple models, we’re doing much more than 500,000 units a year with them. But from the three customers that John talked about, it’s 150 to 500 per year is kind of what we’re looking at.

I assume that they’re new products and they are going to be launching as new ones, and assuming some of these with the ultra laptops you’re talking about, scores of thousands of units, generally in a pre-launch sort of mode. So if you back that out for fourth quarter, that they would seem to be well over 100,000 for each of those. Is that the sort of magnitude and, again, Carl, are these indications, or are they firmer orders for delivery in more the fourth quarter, period?

There’s no orders—sorry, that’s not true. The first product, the GPS system, there is an order for that. And that’s the most solid of all three. Actually, there’s pre-production—there’s production orders for the GPS system, there’s pre-production orders for the smart phone, which is what Carl commented on. And we’re, we’re still really at the sampling stage for the UMPC, but they’re moving very quickly. So they claim to be going into production at the end of this quarter. Whether they’ll make that or not, I don’t know. We don’t know these guys, like I said. We don’t know how realistic they are in terms of when, of how quickly they can move. We only know what they’re telling us.

And last thing, I know that you guys are somewhat limited in terms of what probably your lawyers like to say, but Carl, can you give an indication at least in terms of mix, whether you were on the higher side or lower side of the original margin kind of range? If you could even give that sort of direction.

I can’t, because of the stock option activity, we couldn’t give, if you will, non-GAAP numbers without giving GAAP numbers, and we’re not quite there on our GAAP numbers yet.

Okay, thanks. Thanks, guys.

And Eric Glover with Canaccord Adams has our next question.

Thanks. I was just wondering if you could, this is a follow-up on the last question, actually. If you could give a little bit more color about on things, you know, what percentage in ballpark terms of these new design opportunities are you actually converting at this point to actual design wins?

Well, the three examples we gave are design wins. The three that I told you about, the customers have told us we’re going to be on the shelf by Christmas. And we’re already being rolled out of the GPS system, so the other two were not on the shelf yet, and I said we’ve already taken pre-production orders on the smart phone. And are still engineering sampling the UMPC. I don’t, maybe I don’t understand what you ask, Eric.

No, that’s okay. I think I get the message. It just seems to me that the rollout of the new product which we expected to hit pretty hard in the third quarter has been substantially pulled back, and I’m trying to figure out what exactly is going on. Is it a market issue? Or, I’m just unclear on that.

Well, I think part of the issue is that the engineering folks, we have found in these companies, tend to be more aggressive about their schedules than the reality. And I think that’s part of learning how to do business with these guys. We’ve seen numerous cases now, where the engineering guys have told us that they’re going to be in production a quarter or two quarters before they actually are. So, like I said, we don’t have a lot of experience, we don’t have any experience with these high-volume consumer guys. And so we, we may have screwed that up.

Okay, that’s it for now. Thanks.

Okay.

And we’ll go with Mr. Lawrence Borgman with Jesup and Lamont.

Yes. Can you, in terms of your European customer, which is going quiet this quarter, is there any chance that you could be designed out of that product? It is an interface product that your function might be included on another chip?

Not to our knowledge. They’re continuing these parts, there’s three different cards, and these parts are being done with the same—it’s a Qualcomm chip set, actually, and they’re being done with that same Qualcomm chip set. So, no, that’s not going to happen. What could happen, which hasn’t happened, but what could happen, is that subsequent parts for a higher speed, HSPPA and HSUPA, 3.6 megabits, 7.2 megabit, we could be frozen out of those designs. That’s not determined yet. But it’s extremely unlikely that we’re going to be replaced in this design, or designs, there’s actually three of them.

Okay. And in terms of your customer inventories of your parts, what sort of shape are they in? How are your pricing negotiations coming along? How was pricing last quarter?

Well, I guess from an inventory perspective, these parts are being built, generally being built, these products are being built by contract manufacturers. And I think they learned their lesson about, these were all custom parts, by the way. You know, we programmed them for these customers. So I think they’ve learned their lesson about buying inventory in advance of their customers giving them solid requirements. So I don’t think they’re taking any inventory risk at all. In fact, the way they jerk around and expedite as we move through the quarter for quantities, would tend to indicate that there’s not a lot of excess inventory there in the system. I think with regard to pricing, any of these high-volume customers, in the semiconductor industry there’s always a pricing pressure, and so that’s why you’ve got to keep, you’ve got to keep innovating, you’ve got to keep reducing costs, which we’ve done a lot of.

One of the nice things for us is PolarPro™. It’s twice as dense as Eclipse II™, and for many of our customers, that can be a cost reduction path. So we think we’re in pretty good shape from a pricing perspective, Tom, isn’t that correct?

Yes, we’re mainly, our parts engineering people, our ops people, are very aggressively going after costs, which we’ve never—having been an FGPA supplier, supplying people tend to, 25,000 pieces a year was not something that was our central focus. You start talking to people who are buying a million of something, man, a nickel makes a big difference, and they’re working you over for it all the time. So we’ve done a lot of things to really radically reduce cost. This is an example, where we’re bringing on multi-die programming, so we’ll be programming, we’ll be, actually we’ll be probing and programming dies 16 at a time, which cuts the cost significantly for probing wafers and for programming the devices. So those are the kind of innovations that we’re looking at and really driving to scrub the cost out of our products.

And really, we think this is really one of those dynamic markets where, if we can offer lower prices, our share goes way up.

Okay. So, there’s basically no reason to think that the types of profitability seen in the past is going to, or has changed in the last quarter, or would change in the near quarters?

Well, we’ve always been challenged. Like I said, whenever we’ve had large customers, you’re always challenged to lower margins. But the fact of the matter is, we’re not leaving our old customers behind. So some of this new business is going to be at lower margins, just like all our higher-volume business in the past was at lower margins than the lower-volume business. But the military business, the industrial, the instrumentation test folks, those margins are still good. Actually, those margins will improve as more of them move to PolarPro™. So, our model, we believe our model is still intact, and haven’t modified that, and that was at a 25 million run rate, a quarter-to-quarter run rate, we’d be between 60 and 62 points of gross margin.

Okay. Just one final question. The microcomputer you were talking about? How small a machine is that?

I don’t know if you’ve seen these UMPCs. They’re about one-third the size of a laptop. Actually, Larry, there was one where we had breakfast in New York two weeks ago, there was one at the lobby bar there that Samsung had put out. That happened to be an X86-based, but the ones that are going to be very low—that cost about $1,000—the ones that we’re talking about are very low cost, clearly sub-500, and even lower cost than that. And are not X86-based silicon. So it’s a 7-inch kind of screen and very lightweight.

Okay. Thank you.

Paul McWilliams from Indie Research has our next question.

How you doing, Paul?

I’m doing very well. Thanks for taking a minute to speak with me. Back on the ultra-mobile computer, or the PC, I was doing a little Googling while I was waiting for you there and saw a few of those come up. What other processors are they using besides the X86?

They use Xscale. The original—wait, there’s some confusion going on here, because originally, the term “UMPC” was controlled by both Intel and Microsoft and was architected originally around the X86. And so there was a fairly rigid definition of what UMPC meant. And that’s kind of more now into being processors that can run, may not be running Microsoft, and may very well not be right at X86 architectures.

Gotcha. Are you doing a hard disk control on this application?

Yes, I can’t talk about what we’re doing there, but that’s a good bet.

Okay. Going back and just crunching some numbers here, and I understand that we’re not looking for 2006 data, it’s not 2005 anymore. If I go down to 90% of 2005 and base it off the first two quarters plus guidance, middle of guidance, I’m seeing an implication that we’d have to hit about $44 million in Q4 to hit 90% of 2005.

I think you’re a little bit off on that, but—

Okay, help me out.

Well, with stern 18.6 through the first half, we’ve gotten 12.9 for a guidance, right, so we’ll get 27.5, so to beat next year we’d have to do roughly 21.

Oh, gotcha.

But if it’s a good year, conceptually you’re right. It’s still a big number.

Do you see something out there potentially in the high teens, even?

I can’t do that.

Okay.

One quarter is all we can, and that visibility is really a stretch until we get a better feel for these folks.

Okay. I understand. I’ve seen these situations before, and I know what you’re going through.

You feel my pain, huh?

I wasn’t going to say that. Well, I think that that’s really all I can ask you right now.

Okay. Thank you, Paul.

Thank you.

And next we’ll go to Dan Scoville with Tokimeki Research.

Good afternoon, Dan.

How are you, how you doing? A couple of questions here for you. You said you were just overseas there, and obviously your business is through the ODMs. Do you know who the OEM and what the brand is on these systems?

On some we do, and others we don’t. On some it’s multiple OEMs. And some of the ODMs do their own OEM-ing. Sorry, do their own reselling. So there’s a mix of all of that.

Okay, so are they, for example, the GPS. Are you talking one model or are you talking multiple models, or maybe one model with multiple brands, or multiple models with multiple brands?

Initially, we’re talking one model with multiple brands.

Okay. And that’s in general true across all three of those applications?

No, if you look at how those guys work, the ODMs work, many of them supply variants or skews. They might supply 40 different skews off the same basic platform. And they may make, and that’s the whole deal, maybe make 150,000 to 500,000 of the same skew, but they’re making 40 different skews.

Do you have any insight to the number of skews per platform?

Not on our products yet, no.

Okay. Do you have an idea of when you might be able, the OEM and brand [inaudible] might be able to be disclosed with your involvement?

You know, the challenge is, I asked this question when I was there, because all the investors always ask, “When are you going to be on the shelves?” and “Whose products are you going to be in?” And all of the ODMs were under NEAs with all of them that we can’t discuss that we’re doing business with them, and we can’t discuss which products that we’re in. So it makes for a challenge. What happens eventually, though, I’ve noticed with other people’s product, is that it kind of leaks out into the market. And typically it doesn’t come from the guy who’s making the silicon, but comes from other people. The thing that’s strange about this is that everybody in Taiwan knows who’s in what part, but those guys certainly don’t want you talking about it.

Okay. Okay. Shifting gears a little bit, clearly, the PolarPro™, I guess you’re actually starting to roll that out now. Is there a bad-news story, is it maybe cannibalizing some of your Eclipse II™ opportunity, or is it still kind of far enough away that it’s really not a problem?

Well, I think it had the potential, but I’ll tell you, Eclipse II™ was so good from a low-power perspective, that we announced PolarPro™ in November, and we thought we’d get more folks who were in design do a design conversion, and they didn’t. And the reason they didn’t, first of all was, obviously, Eclipse II™ was a bird in the hand and well understood, and they’d been in production a while. And, of course, PolarPro™ had not. But the other was that they were getting the results they wanted, and these folks are very loath, once they start down a path for a given project, they are very loath to make changes midstream. And so I think what we’ll see is we’ll see an uptake on new products, new projects, for PolarPro™, and there it will be around being able to offer them cost reductions and still keep good margins for ourself.

Okay. When you, I’m trying to figure out this new customer base of yours. Could it be characterized as simple as you’ve got to focus, if you don’t have a PO on this, so you’re trying to figure out how to balance those, too?

No. You know how these guys work. You never see POs. Nobody orders stuff six months in advance anymore. So you’re always, typically three months is a long time to have a PO. And none of it’s firm beyond 30 days. So people don’t buy semiconductors where they give you any kind of visibility. Anybody who, and our business, obviously, has always been a very trends oriented business. We’re getting pretty good visibility and pretty good predictability out of the European guy, but that’s, we’re three quarters into that deal at this point. So, we just don’t know yet is the real story, Dan. And I’m sure as we get closer to these guys and become a more valuable contributor to them, that we’ll get better insights, just like we are with our guy in Europe.

Okay, let me ask you a slightly different way. You’re not even getting a forecast from this, you’re not sure how to judge yet?

Well, you get a forecast from the purchasing people, that, “We’re going to start production in this quarter, and it will grow at 10 and 20 and then 40K units per month,” and then building to, “We’re not quite sure.” So that’s the kind of visibility you get. And then that month comes around, and the 10K or the 20K that they were going to buy may or may not materialize or, in the case of the other guy, in Europe, it was going to be 30 between then and the end of the year, and 60 between then and the end of the year, and it wound up being damn near 400. So, you tell me how to forecast something like that. The only good news is that we’ve got, we’ve bought this last quarter, I think Carl told you, we bought guaranteed capacity, guaranteed assembly capacity, we’ve got guaranteed wafer capacity, we’ve got plenty of test and programming capacity, and so we could have some significant upside and be able to respond to it.

Good, good. No more questions. Just kind of the risk of beating a dead horse here. Your ESPs are softer than expected, I understand in relation to the p1 and the p2, and I understand the run-up is getting muddy here on the new stuff, but the ESPs, that—some of the reason for the softness here, some of us were thinking that that was going to be fairly steady business.

It is pretty safe for us, Dan, and we just saw a softening in bookings in Q2, our sales guys think it’s temporary, but by the same token again, and Q3 is really a tough quarter for us from a bookings perspective, primarily because of European vacations. And so we’re just, we’re being cautious, we’re not being real bullish about the immediate rerun in that booking rate. Do we think that has potentially gone away? No, but is it going to come back necessarily really strong in the third quarter? We don’t think that either.

To have that much knowledge, it sounds like it’s pretty narrow in terms of its softness.

Well, it was a couple of, there was more than a few people that were soft in the quarter. But those same people are forecasting, are forecasting they’ll come back, just not to the levels that we were expecting a quarter ago.

Okay, thank you, guys, again, and good luck, there.

Okay.

And it appears there are no further questions at this time. Mr. Hart, I’d like to turn the conference back over to you for any additional or closing remarks.

Well, thank you, folks. We look forward to getting back on track here as quickly as we can relative to our financials, and we appreciate your interest and hanging in there with us. We’re giving it the best we can. Thank you folks. Bye-bye.

Thank you. This does complete today’s conference. We appreciate your participation, and you may disconnect at this time.