CEVA Inc (CEVA) 2004 Q1 法說會逐字稿

Thank you for standing by. Welcome to the CEVA first quarter 2004 financial results conference call. At this time, all participants are in a listen-only mode. Later, we will open up the call to your questions, and instructions for queuing up will be provided at that time. As a reminder, this conference is being recorded. The conference call is the property of CEVA, and any retransmission, or copying thereof, is strictly prohibited without express written consent of CEVA, Inc. I would now like to turn the conference over to Deborah Stapleton for introductions. Please go ahead.

Thank you, operator, and thanks to all of you for joining us today. On the call today are Chet Silvestri, CEVA's President and CEO; Christine Russell, the company's CFO; and Barry Nolan, the Vice President of Marketing.

Before we begin the presentation of the first-quarter results, I want to remind listeners that any statements contained in this conference call, including the Q&A period, that are not statements of historical fact, may be deemed to be forward-looking statements. Including statements with respect to CEVA's anticipated financial results for the second-quarter fiscal 2004, and future quarters.

Words such as expected, believed, anticipates, plans, expects, will, and similar expressions, are intended to identify forward-looking statements. There are a number of important factors that could cause the results of CEVA to differ materially from those indicated by these forward-looking statements, including, among others, risks detailed from time to time in the company's SEC reports, including its annual report on Form 10K for the year ended December 31, 2003. CEVA does not undertake any obligation to update forward-looking statements.

Now, having said that, I'll now turn the call over to Chet Silvestri, the company's President and CEO.

Thank you, Deborah, and thank you all for attending the CEVA first-quarter conference call. Today I'll discuss recent highlights of our business, and our progress in extending our position as the leading licenser of DSP and communications technology to the industry. After I complete my remarks, our CFO, Christine Russell, will then discuss the financial results.

We completed our first quarter with revenues of $9.2m, and a profit of $409,000, or 2 cents per share. This is within the range of guidance we gave, and demonstrates that we were able to achieve a significant goal of the fourth-quarter realignment, profitability.

We completed five new licensing agreements in the quarter, three for our DSP technologies, and two for our GPS, or global positioning system technology. In addition, we continued our progress on a second key goal, extending our DSP leadership. Here industry adoption is a key indicator of our success.

Last week we were very pleased to announce that Samsung had become the first licensee of our new CEVA-X product for their next generation 3G handsets. Samsung is a leader in the mobile industry, and the Samsung adoption of CEVA-X is a tremendous endorsement for us.

Also we have announced yesterday that EoNex, one of Asia's most progressive wireless players, has licensed CEVA-X for its next generation 3G wireless chipset. We now have three licensees for CEVA-X since its introduction in late Q4.

In the quarter we also announced that technology leaders such as Kawasaki, Oki, Stepmind, and National Semiconductor have licensed our technologies.

Geographically, we are continuing to see strong business growth in Asia. Our focus on adding sales, marketing, and business development resources in Asia has paid off. Asia has moved from being less than 15% of our revenue a year ago, to comprising 35% of our revenue in this quarter.

I've just returned from a business tour of Korea and Japan, and saw strong interest in our latest technology, particularly CEVA-X and Xpert GPS.

CEVA licensees continue to be successful, and to gain market share in the DSP market. Their combined semiconductor shipments containing our technology were up 9% sequentially, to 19m units in the quarter. Looking forward, we expect annualized CEVA license shipments of approximately 80m units in 2004; an increase of over 45% compared to 2003 buys.

The industry in general continues to grow. Last week Forward Concepts, the preeminent industry analyst for DSP, announced that final 2003 DSP revenue from the World Semiconductor Trade Statistics Group, was $6.1b, representing 26% growth over 2002. Consumer shipments, in dollars, grew 109%, and wireless shipments grew 32%. The combination of the wireless and consumer segments represented approximately 70% of that total dollar volume in 2003. These are the primary markets that we are targeting with our products. Forward Concepts is forecasting that 2004 will see 25% growth in total DSP shipments.

As the only meaningful independent source of leading-edge DSP cores and solutions in the industry, CEVA is uniquely positioned to serve this growing market need. Our new product programs are on track to deliver on the DSP opportunity. The CEVA product strategy is focused on three areas: first, DSP cores; second, the framework for system on chip environment around these cores; and third, application, or full system solution, such as communications, multimedia, and GPS solutions that are powered by our DSP core.

In the quarter, we continued our aggressive ramp in our DSP portfolio, with the launch of two new application programs, Xpert-Blue, a complete Bluetooth 1.2 solution, and Xpert-Media, a complete audio and video platform for both the cellular and consumer electronics markets. We've already secured two licensees for our Bluetooth platform, which is the first licensable Bluetooth standard version 1.2 in the industry.

Xpert-Media, launched in February with planned shipment beginning in Q2, is a particularly exciting technology for us. It targets DSP's highest growing market, digital multimedia. Through patented techniques and innovative utilization of our DSP, we deliver audio and video multimedia processing in an all software solution, with performance and cost superiority over what is achievable in hardware today.

In the second quarter we'll continue this aggressive new product rollout. In a few weeks we'll be formally launching our Xpert-GPS platform, which is the eighth generation release of our GPS location technology. Uniquely, this platform is powered by CEVA DSP and is delivered as a largely software-based, configurable product. We have already signed two licensees in the first quarter for our Xpert-GPS.

Also in the second quarter, we'll begin shipment of our voice over Internet protocol, or Xpert-VOP platform. VOIP is forecast to be the largest growth area in wired communications.

Binding all these applications to our DSP cores is a complex and challenging task for our licensees. To that end, we'll be delivering a major upgrade to our Xpert open framework environment. This release will deliver a standardized software framework, enabling fast and efficient DSP application developments and integration.

Partnering is also key to our strategy. Earlier in the quarter we announced the establishment of a foundry program with UMC, for our Teak and TeakLite DSP technologies. In the second quarter we expect to add additional partners to our foundry program. Foundry programs are an excellent channel for delivering a rapid and lower cost of entry for fab-less organizations looking to harness CEVA DSP technology.

We are also exploring new channels for our non-DSP-centric IPs, namely our serial ATA and PLL IP. Both these technologies are well-placed to capture significant traction in the industry. We feel that by adding effective sales and marketing channels, we can accelerate this growth.

In summary, we are on target with product releases and financial performance. I'm pleased with our DSP growth and adoption in the first quarter. The DSP and communications markets remain among the fastest growing markets in the industry, and this, supported by our new product programs, continues to give us confidence in achieving our corporate objectives of growth, profitability, and DSP leadership.

I'll now turn the call over to Christine Russell, who will tell you about our financial performance.

Thank you, Chet. For the first quarter we reported revenues of $9.2m, and a profit of $409,000, or 2 cents per share; both figures in line with our guidance. As Chet mentioned, we are particularly pleased to report a profit, as being a profitable company is one of our key goals for 2004.

Revenues for the prior quarter were $9.6m, with a loss of $9.6m, or 53 cents per share. This loss included $9.1m of restructuring charges as we realigned our business to be in line with our long-term objectives. In the first quarter we have no restructuring charges, and looking forward, we don't anticipate having any further restructuring charges related to our fourth quarter business realignment.

While our total revenue for the quarter is slightly lower than the previous quarter, on a comparable basis, we're pleased that we replaced a good portion of the revenue from exited hardware-related product lines with growth and revenue from our continuing licensing business. The exited product lines were contributing about $1m revenue per quarter throughout 2003.

We mentioned last quarter that we would no longer be generating revenues from these exited products, starting in the first quarter of 2004, and looked forward to replacing at least some of these foregone revenues with growth in our key business, the DSP-centric products.

So, for example, on a comparable basis, if we subtract $1m of revenue from our fourth quarter, which was $9.6m, we get $8.6m of revenue from the ongoing businesses. Our revenue of $9.2m in the first quarter represents a 7% increase over this adjusted comparable fourth quarter. This comparison demonstrates that we are achieving our goal of replacing these foregone revenues.

Gross margins were 84% compared to the prior quarter’s gross margins of 85%. The 1 point reduction was due to the slightly lower revenue contribution from royalties. Royalties yield 100% gross margin, and licenses generate a slightly lower gross margin than royalties.

Looking at the balance sheet, we closed the quarter with $59m cash, with $500,000 used in the quarter. The primary cash used was for capital outlay for design tools. DSOs were 105 days; this is slightly higher than our 2003 average of 91 days. Since the majority of our revenue is from licenses rather than royalties or services, we continue to see a majority of our revenue back end- loaded in the quarter.

As in prior quarters, the high DSOs are a result of the typical non-linearity of the quarter. We ship around 2/3 of our sales in the last month of the quarter, making it difficult to collect on those sales intra quarter. Quality of receivables is not a problem, with 75% of the accounts receivable under 30 days.

The revenue mix was 6.6m licenses, 1.2m royalties, and 1.4m services. The predominance of the license revenues is a predictor of future royalty payments, as our licensees successfully come into production with our technology in their products. Licensing revenue remained the same quarter-over-quarter. This was expected, due to the loss of revenues from some of our exited businesses. As I mentioned, we're pleased that we were able to replace quite a bit of that foregone revenue with new business from our ongoing products.

Royalties were $240,000 lower than the prior quarter, primarily due to a major customer moving through the volume shipment traunches to a lower percentage rate payment. The average cents per chip paid in royalties was 6 cents in the first quarter of 2004, compared to an average of 7 cents for 2003, and 8 cents in the prior quarter. The lower per chip amount again relates to customers who moved into volume production paying a lower amount per chip.

While we expect royalties to grow over time, as our customers enter production and pay us per unit royalties, we expect the royalty revenues may be lumpy rather than linear, quarter-over-quarter. This is due to a couple of factors. Customers generally pay royalties with differing percentages as they achieve volume shipment goals. The percentages generally vary from 6% for initial shipments, down to 3% for high volume shipments. Customers may also enter the royalty paying production phase while others exit it, as they wind down production on individual products.

Our top three to five royalty payers typically represent as much as 60% of the royalty revenues. This is because we are still at an early stage in our royalty revenue franchise, and the revenue stream is in the building stage. This concentration of royalty payers can also be a factor in quarter-to-quarter volatility of royalty revenues.

We had 18 royalty paying customers in the first quarter, up from 17 in the fourth quarter. Our licensees reported a total of 19m chips shipped, up 8% from 17.5m in the previous quarter.

Our fourth quarter work to make sure the business is cost-effective paid off. Our operating expense was $7.4m for the quarter, which is over $1m quarterly expense reduction from the fourth quarter of 2003.

Looking ahead, we expect moderate growth in revenues for the second quarter of 2004. We expect gross margins and operating expenses to be stable. For the year 2004 we continue to target revenue growth up 10% or around $40m. While growth of the ongoing DSP-based business is targeted approximately in line with market growth at 20%, we do have to factor in the elimination of approximately $4m of annual revenue from the discontinued non-strategic products in our year-over-year top line growth comparison.

We continue to target profitability for 2004. The average share count for 2004 is expected to be approximately 21m shares. We'll now be happy to take any questions.

Thank you. At this time, if you would like to ask a question, please press star, followed by 1, on your telephone keypad. Your questions will be taken in the order in which they are received. We will announce your name prior to asking your question. Once again, to ask a question, please press star, followed by 1.

Our first question comes from Gary Mobley. Sir, please announce the company you represent.

Hi, Gary Mobley with B Riley & Company. Good morning. I just had a question, Christine, on the operating expense guidance for Q2. I assume you're expecting some sequential increase from the March quarter as well. But, is it consistent with the moderate revenue growth you're forecasting for Q2 as well?

Well what I said, Gary, is I expected operating expenses and gross margins to be stable, compared to the quarter that we just reported. So that would say that operating expenses would be not be increasing, which I think is in line with saying that we'll have moderate growth on the top line.

OK, I missed that. What was the employee count, at the end of the quarter?

We had 217 employees. I'll break it out for you. 160 of those were in R&D, 20 were in Sales and Marketing and 37 were in Management and Administration.

OK. And, other revenues were a bit higher than what I'd forecasted, especially in light of the $1m elimination in Hard IP business. What was the reason for that?

The reason for that is we had some design services contracts that went into that line.

OK. I'll open it back up to other questions. Thanks.

Thank you. The next question comes from Matt Robison. Sir, you may go ahead and please announce the company you represent.

I'm with Ferris, Baker Watts. Good morning, congratulations. Chet, you mentioned those Xpert-Blue licensees. Were those already this quarter? Or were those before the first quarter? You counted two GPS and three DSP for the first quarter.

No, the Xpert-Blue 1.2, so our latest release of GPS went out this quarter. Prior to this, we'd been shipping 1.0.

You mean GPS, or Bluetooth?

Bluetooth.

OK. So, did you say that you had two licenses, already of that Xpert-Blue 1.2?

I think I said two licensees for GPS and three licensees for DSP.

Yeah, I heard that part. I guess I heard 1.2 and I thought you said the 2 was two licensees, OK.

Oh no, sorry. 1.2 was just the latest release of our Xpert-Blue.

1.2, sorry Gary. It's Barry here. 1.2 is the Bluetooth standard, the latest (indiscernible) of their standard, they call it Version 1.2. So, our platform has been upgraded for that standard. We've done some licensing already in that platform. But as Chet mentioned, there are also two additional licenses in Xpert-GPS, which will be formally launched in the coming weeks.

Oh, OK. So, two additional -- I think I've got the Bluetooth figured out. Now, I'm thinking you're talking about two new GPS. Or, is that two new GPS in the March quarter?

Correct, two new GPS.

OK, so you didn't talk about any licensees so far this quarter?

No.

OK. All right, I hope I don't have everybody else on the call confused now. Anyway, would you say, Christine, that the seasonality in the consumer electronics-oriented Semiconductor business, in that, you know, big shipments towards the end of the third calendar quarter, would you say that that is less of a factor on your royalties than the pricing agreements in the volume thresholds, that any given customer might make, say, in a quarter?

No. All of those things are factors, but, as I said, we do have somewhat of a customer concentration. So, any customer moving through the volume traunches that we've established for them, from 6% down to 3%, is going to lower our overall revenue, although we're very happy to see our customers be very successful in production. It bodes well for the future. But, I would say that is the primary reason, rather than seasonality.

Did you see any shifts, in the quarter, to/from cellular disk drive? You have major royalty applications. Was there any one that was particularly strong, or particularly weak?

No. Wireless and multimedia were the predominant royalty pairs.

OK, and how's the pipeline looking for the current quarter?

You know, we don't, in general, talk about the details of our pipeline. But, let's see, what can we say about this?

You can talk about Asia.

I think, as I mentioned in Asia, where I was last week, we saw a particularly strong interest in CEVA-X and Xpert-GPS. And I think those two elements are going to become a growing part of our sales forecasts in pipeline.

You had eight new licenses in the fourth quarter, five in the March quarter. Is that budgetary seasonality? Should we expect to see something come back towards the eight kind of level this quarter?

No, I think there are a couple of factors. One clearly, as I think you heard, our businesses ramped up substantially in Asia, and the first calendar quarter in Asia is a very bad budgetary quarter. It's the end of everybody's fiscal year. They're all out of money. Their new fiscal years all start April 1, so that is a big factor.

But the other factor is, our newest technology, as I said, our Xpert-GPS and CEVA-X are very high-ticket items. Their ASPs are substantially higher then our prior cores and technologies. And so, we'll be working on fewer, bigger deals, potentially, in these product categories.

Yeah, I mean, because the applications licenses really haven't changed that much in price. It's the DSPs that have gone up. Is that right?

DSP and our latest generation of GPS, which we call Xpert-GPS, which used to be called the 4000 Series, will have a premium price attached to it as well.

So you obviously don't see the normal year-end as a budget flush? You see it as a budget constraint, it sounds like.

Yes.

So, you think things will be a little healthier in the June quarter then?

Well, Asia as a region. The rest of the regions don't have that kind of seasonality.

You mentioned Asia for '03. How about just for the fourth quarter of '03, how much did it contribute?

I don't know if I have that in front of me right now, Matt.

OK, I'll yield the floor for now.

Thank you. Our next question is a follow up from Mr. Gary Mobley. Sir, you may ask your question, and once again please state the company you represent.

Hi, B Riley and Company. How much was Europe down, sequentially, from Q4? And, what do you attribute the decline to, and what, potentially, would be the catalyst to re-ignite the activity in Europe?

I'll take that question. Geographically, Europe was actually up in the first quarter. But the interesting trend that we're seeing is, Asia has steadily increased for the last, say, four or five quarters, whereas, the U.S. and Europe seem to be trading places as being the predominant remaining two regions. Since a lot of our deals are very large-sized, one large deal in Europe or the U.S. can really swing that one way or the other.

So, in this quarter, let's see. Geographically, Europe was 45% of the sales, the U.S. was 20%, and Asia was 35%. But again, while we expect to see Asia continue to increase, Europe and U.S. may trade places, from time to time.

OK. Nothing really the reason that it’s there?

No, I would not read a lot into it.

OK, and Chet, on a competitive view, we're hearing that LSI Logic is trying to garner some share for its CSP family of cores. Is that creating a pricing pressure for you guys, either at the low end or the high end? And, more importantly, is it creating any pressure when you negotiate royalty rates?

We're not seeing any pressure overall in our royalty rates, anywhere. I think there are a lot of competitors in the low end, you know, where our Teak and TeakLite compete. We've been getting the lion's share of that business. We did last year, and we expect to continue to do that. But this is where LSI and a few others compete.

In the CEVA-X, where our newer deals and our larger deals are coming, we have no competition there, no pricing pressure. And, we're going to continue to push very hard for penetration on these new technologies.

OK, and Christine, tax rate for the full year, what's your best estimate?

Somewhere between 10% and 15%, so to be conservative, I'd use 15%. We have the advantage of having a 10% tax rate in Ireland for any of our software IP, and a 15% rate in Israel for any of our DSP technology.

Now, I wouldn't average the two, because we also pay some franchise taxes and taxes in Delaware and other miscellaneous taxes. So, 15% would probably not be a bad number to use.

OK, and what's your best estimate for a blended royalty rate per unit, for the full year? In other words, you know, you're trending down here in Q1. Will there be significant royalty pairs coming on that are going to, perhaps, change the tide and increase the rate? And, you know, where might we end by the end of the year?

Gary, it's Chet. I'll take that. I think as we said, the CEVA-X licensing structure is at a premium to our old Teak and TeakLite, both in terms of upfront license fees and royalty percentages. So we expect the trend, in some way, to go up. But, we really haven't done the work, I don't believe, to come up to really have an accurate or, you know, realistic understanding of where it's going to be this year. We've got a feel (ph) that we're just starting to ship these technologies. We aren't yet comfortable looking at what the volume growth is going to be, and where the royalty payments will start kicking in. And, I don't think we’re going to know that until more toward the middle of the year.

OK, all right. Thanks again.

Thank you. Our next question comes from David Foundry [ph]. Sir, you may ask your question and please state the company you represent.

Yes, Hartman Funds. Good morning. Matt did confuse me on the licensing agreements, and I'm sorry, but, you referenced that there were five new licenses within the quarter. Does that include Samsung and EoNex? I mean you could interpret that sentence as those happening after the end of the quarter.

Sure, so we had five new licensees in the quarter. EoNex was one of them. We just put out that release, I think, within the last few days. Samsung was announced in the fourth quarter.

OK, so both —okay. Sorry.

We couldn't announce the name in the fourth quarter, because of confidentiality commitments with Samsung.

It was included in the eight in the fourth quarter?

Correct.

So, EoNex was in the first quarter? And then, the last sentence in that paragraph was, "Further industry adoption behind CEVA was evidenced by licensing announcements with Kawasaki, Oki, Stepmind, National Semiconductor." Are those the other four that were recorded in the -- ?

Hi David, Barry here. No, just in -- there's a difference between deals signed and deals announced. Typically in the IP industry, you normally do a deal, say, in this quarter, and you typically may not be able to announce that in the second quarter – (MULTIPLE SPEAKERS)

OK.

-- deals done and deals announced. Oki, Kawasaki, etc., were all deals done in 2003. They're announcing today because our customers have allowed us to announce them, either because they've taped (ph) to us that the product is ready. So, I mean, there are two things. Five deals in the quarter, three for DSP, two for the GPS, and then there are a number of announcements, which is more of a marketing solution.

I think you can -- if you relate apples to apples, we've really only announced one of the five deals for the fourth quarter, which is EoNex. The other four remain unannounced at this point.

OK, that helps a lot.

Sorry for the confusion.

No, that's quite all right. And then, on the royalties side of matters, you indicated that one of the customers moved from a -- moved volumes up high enough that the royalty rate dropped. Do you then, is it just the incremental units that you sell on, where the royalty rate drops? Or, do you actually go back and adjust all prior royalty payments?

It is not the incremental units. It is all the royalty units for that quarter. So, it doesn't go backward in time. There're no credits given.

That's what I was trying to get to. So there are no credits given?

No, but for example, if a customer shipping a million units in -- I'll give you just a hypothetical example. If the million units in Q4, are at 10 cents a unit, and then pass the threshold of a million, and then the price goes to 5 cents a unit, if they ship the same million units in Q1, they would only pay us 500.

Got it, I got it. OK, so it doesn't go back and adjust down so that, and your example, you could come up with zero revenues in Q1?

Yeah, that can’t happen. The actual revenue could be lower, even on higher volume.

And then, can you give us some sense, on this, you've now announced the Samsung. When would you anticipate starting to receive revenues from that licensing agreement?

Generally, for a major DSP licensing agreement, it takes between 12 and 18 months for those products to be shipped into the market, with some reasonable volume. I'm not talking about early Samsung .

Right, yeah.

So, it's a 12 to 18-month horizon, generally.

So, that was, in the fourth quarter, we might expect something, some limited revenues in the fourth quarter of this year, or maybe a little bit longer than that?

From a royalty, yeah barely, because remember, we're one quarter in arrears. And that pushes you more toward the 18 months.

OK. Thank you very much.

Thank you. Our next question comes from Rich Valera. You may go ahead with your question, sir, and please state the company you represent.

Thank you. Needham and Company. Hi, I was wondering if you could talk about the competitive landscape in the GPS area, how your Xpert-GPS product compares with dedicated hardware solutions from Surf, or Qualcomm?

Sure Rich, it's Chet, I'll make a few comments, and maybe Barry wants to add to that. But because we are selling GPS intellectual property and not chipsets like from a company such as Surf or with Qualcomm, which is (indiscernible) as well. We are primarily targeting our sales to those companies that want to integrate GPS as part of an SOC, system on chip solution, where they need to put it as part of a bigger framework. And in particular, our target market right now is the cellular handset makers, where we have already a very strong position with our DSP cores, and now we've provided a GPS add-on that is powered by our DSPs, that they can simply integrate into the cellular handset chipset, potentially all on the same chip as an application processor.

And this is where we have a unique advantage that you're not going to be able to buy individual chips from another company, such as Surf, or you have to create a module for it; you're not going to be able to create an SOC out of what they give you. So we're targeting sort of a specific piece of the GPS target market, in particular cellular handsets, and to some extent automotive, where integration is an important part of it as well. But if integration is not important, we are not going to prioritize that very high in our efforts.

With respect to performance, can you put up comparable performance, to say, a dedicated Surf GPS chipset? Do you have a sense of how you compare performance wise?

We do. We have, and some of our customers have, done detailed competitive comparisons. This is the other distinguishing feature of our offering. Our offering was designed, and all this stuff comes historically from military applications, ours has two unique aspects to it. One, ultra low signal strength capability, what we call so-called indoor GPS. So if you lose direct sight to the satellite, we can still maintain a tracking, and not by using a network assist in a cellular framework, like some other competitors try to do. But we can literally, with our software and our hardware and algorithms, maintain indoor tracking with GPS.

The second major feature is very rapid acquisition time. Some of the consumer handheld GPS units that you might buy for $99 can take up to a minute to locate the satellite and do the computation, and give you a location. Ours takes seconds, and this is very important as well when you're traveling, when you're going in and out of satellite contact, you have to maintain a very rapid updates, because you might have only a few seconds to get a fix.

So these are unique, we are very premium featured GPS. You're not going to see us in the $90 hiking GPS unit probably.

Can you give us a sense of how powerful the application processor has to be, to be able to run your Xpert-GPS solution? How prevalent do you think application processors that powerful will be, in general, throughout the handset world?

Sure. So, the way our GPS has been designed it is mostly self-sustaining. You need some computation power from a DSP to compute the locations, but DSPs are very powerful computation engines, that's what they do. It takes very little power. We’ve talked about using DSP resources in terms of megahertz, and it takes only a few 10s of megahertz of a DSP’s power to run the GPS. Literally even our ultra low cost DSPs are running 200, 300, 400 megahertz in our customers.

So siphoning off 10 or 20 megahertz is really nothing in the grand scheme of things.

Great. Well that was very helpful, thank you.

You bet.

Thank you. The next question comes from Morgan Franke. You may go ahead with your question, and please state the company you represent.

Hi guys, it's Manchester Management. A couple of questions; one, your design services business looks like it picked up a bunch sequentially. It really basically sort of compensated for all the hard core stuff that you lost. Can you talk a little bit about why there was so much strength there?

Our design services business is, I used the word lumpy before, I'll have to find a synonym for it, but it is. It is not always linear. We take on projects and so that's not always going to be the same from quarter-to-quarter. It's not going to be a steady state business, so we will have spikes in that business. We don't generally talk about individual customers that we do the work for.

But I think it's fair to say our design services business is still alive and well, and we're continuing to take on business.

(MULTIPLE SPEAKERS)

I'm just wondering, is there something in particular that caused a vibrance in that business, the release of new platforms, or why was the uptick now?

The uptick was simply the result of us taking on a project; it had nothing to do with the release of CEVA-X particularly. It's just that we accepted a project and we made our deliverables in this quarter. Generally there are software (ph) deliverables that go along with it, and then the design services are time and materials.

One other thing to add in there, that services line is really two components of revenue. One is consulting, and the other one is maintenance. We've had great DSP pickup last year, particularly in Q4, and that's fed into that line, and that will be sustained in '04.

That is a very good point, Barry. So maintenance continues to be a growing part of our business, based on our DSP base.

Got it. OK, then going back to the two GPS licenses that were announced. Was revenue recognized from those in the quarter, and if so, how did you recognize revenue ahead of shipment?

No, the licensees were for our new Xpert-GPS, which we're not shipping until Q2, were not recognized in the quarter.

We only recognized revenue on the existing GPS that we have, that we did ship.

Got it, OK. So we could conceivably see some license revenue there. Now are those, do you get a big upfront payment there, or is that amortized over time?

We typically get an upfront payment on any kind of a license for any of our products.

OK, but I mean does that all get recognized in one quarter, or will that be recognized ratably over the life of a license?

It's typically recognized in the same quarter, but where there are any deliverables associated with it, any custom work such as the design services we were just talking about, in deliverables, that will typically be recognized over a couple of quarters, sometimes as much as three quarters. That's not an uncommon phenomenon in our business. So we only recognize the revenue on the items that we have shipped and delivered.

OK. Then last question, the 80m estimate that you guys gave for royalty units over 2004, I mean that seems to imply very little growth from where you are now. If you shipped 19m in this quarter, there's not a lot of headroom for you to ship any more in the forthcoming quarters to reach that 80m number. Is there -

Morgan, it's Chet. You're correct; the math that we did in projecting for the year is fairly simple. Now I will say that we do have royalty payers going out of the stream as well. Products get discontinued; they wind down in their life cycle. So there's a life cycle, there's down as well as up, so it's not just always up. But that being said, I think we could expect to see growth in our unit shipments this year.

So perhaps that 80m is kind of a baseline, and if things go right you could do better?

I would say yes.

OK, thanks very much guys.

Thank you. The next question comes from Mr. Sean Murphy. You may go ahead, and please announce the company you represent.

Nomura International. Hi guys. You made a reference to a product set that you’ve yet to launch to enhance the ability of customers to integrate your application products with your core products. I wonder could you give me a bit more information, including how that may affect your pricing model?

Sean, it's Barry here. Just to explain, as far as you see the Israeli DSP core is on a suite of applications around those DSP cores, be it audio, video, GPS, Bluetooth, combining those two together is immensely difficult. We satisfied some of that last year, what we call our Xpert framework, which is the hardware, system and chip environment. We're doing a major release in Q2, which is the software/middleware on top of that hardware environment. So the software/middleware for quick algorithm development on CEVA DSPs and integrating multiple different types of algorithms, video algorithms, audio algorithms and GPS, and simplifying that process.

How it affects our deal value, obviously it pushes it up. I think it's perhaps more getting us into larger portfolios, strategic type deals, where we do license to cores, the framework environment, the complete tool change [inaudible].

Sean, this is Chet. Let me add to that, because I think this is an extremely important point, how our Xpert framework works and adds to our value proposition. As Barry said, let me go back, when we license a DSP core, it is licensed in a form called a synthesizable core, which means that it's easy to port to any foundry or fab process. But at the same time there's work to be done, because you have to add to that core all the IO and all the memory, so you need to put memory, memory interfaces, IO, all kinds of other hardware design has to be done around that core. It takes some time and effort to do that. We have done that, and that was the original layer of our Xpert framework, what Barry referred to as the hardware layer.

So if you license our Xpert framework for our peak core, for example, we give you the memory subsystem, we give you the IO subsystem. We give you all the hard macros, if you will, for that. Now we've taken a step beyond that and added the other layer, which is the software layer, and primarily allow what I'll call a plug and play interface to other application IP from us, or even from third parties, hardware and software pieces. So it's easy to bolt these additional pieces around our DSP and create a full system on chip.

We license it like software, so it's an upfront payment for this Xpert framework. It doesn't typically have royalties associated with it; it has a license fee and maintenance associated with it.

OK, thank you.

Thank you. Our next question comes from Michael McCormack. You may go ahead, and please announce the company you represent.

Hi, I'm with Guilder. Christine, in your comments you said that the $6.6m revenue in licenses was flat sequentially, but only so because of some loss in the hard wire business that you recorded in the prior quarter. Can you give us that number, what was an apples to apples comparison?

No, I'm not breaking that out by the licensing versus the other. I only broke it out on a total revenue comparison.

OK.

The total was $1m in the quarter between the licensing and the hardware, which was in the other revenue.

Right, exactly. OK. But you can't give us any more specific about whether it was 50/50 or something like that?

No, we don't go into that level of detail Michael.

OK. How many royalty-bearing customers were in the prepaid mode in Q1?

Well actually we have eight, and that's been kind of a steady state number. The interesting thing is we have really moved away from accepting prepaid from customers for our future royalty obligations. To give you an idea, in the last six months we've only done two of these deals.

OK. Could you also give us, Chet mentioned some life cycle of the existing revenue that you're generating out of royalties currently right now. How much of your royalty revenue would you say is in product that you would consider to be at end of life type of situations?

Of the top three royalty payers, I don't think any of them are close to their end of life right now. So as I said earlier, we're really pretty early in our royalty franchise. When you think about it, our royalties a year ago in the first quarter, our royalty revenue was about $600,000, and we're now at $1.2m. So you can see that's ramping up pretty dramatically, so these folks are still in the middle of their production cycles, if not earlier.

Barring this one customer that had a step down in their pricing due to volume, would royalty revenues have increased with the exception of that?

Yes, they were one of the three large royalty payers, and they were primarily responsible for the step down in the royalty revenues.

OK, thank you very much.

Thank you. Once again, if you would like to ask a question, please press star, followed by 1, on your touchtone phone. Again, that is star, 1, to ask a question. One moment please. Thank you. The next question comes from Matt Robison. Sir, you may ask your question, and please state the company you represent.

Ferris. This might be under the category of housekeeping, but Chris, you gained, you did better by a penny or so in terms of your net 'other income' versus the fourth quarter. What did you do to achieve that? Was there something structurally changed in For Ex, or however you're treating it?

Actually we've done a great job of managing our foreign exchange, and my compliments to the finance department. What we've done is we formed what you could call a natural hedge. So we've taken a look at our exposure in Euros, since we sell in U.S. dollars, and we made an estimate of the payments that we will have due in terms of Euros, and we've made purchases and put that amount of Euros on our balance sheet, and that has formed a really nice hedge against any kind of FX. As you can see, the results are really paying off. We had next to no foreign exchange costs.

So this is just managing your cash in terms of different denominations?

Absolutely.

Thanks. That's all I had.

Thank you. The next question comes from Dave Hegger. You may go ahead, and please announce the company you represent.

Kennedy Capital. I had a question on the DSOs. You're running over 100 days. Is there anything that you guys can do that might bring that down?

Well, what's going to bring the DSOs down is having more linear quarters, and I think as our revenues tend to increase, we're not going to be quite as prone to being back-end loaded, one would hope. Certainly, as I said earlier, it's not a problem of collections as the quality of the accounts receivable. 75% of the accounts receivable is under 30 days. So, it's really an issue of linearity in the quarter. And unfortunately in this industry, we have trained our customers to wait until the end of the quarter to negotiate deals.

And so, the only cure I see for that is really just having a larger volume of sales so that one or two individual sales don't affect the DSO as strongly.

Let me add to that. The other thing that will affect our DSOs is as the percentage of royalty payments increases, as a percentage of the quarter. So, today we're very heavily weighted toward license fees. More than 70% of our revenue comes from license fees, and those, as Christine said, tend to be very back-end loaded. Customers drag their feet through the quarter.

But royalties and design services tend to be much more linear. And as those become a bigger percent of our business, and we're certainly expecting royalty percentage to rise, that will also have a natural tendency to reduce DSO.

OK, thanks.

Operator, I think we've got time for one more.

At this time, there are no further questions in the queue. I would now like to turn the call back over to your speakers for any concluding remarks.

Thanks, operator. So, I'd like to thank you all again for joining us today, and for your continued interest in CEVA. I'd also like to thank all our employees whose hard work and dedication make our continued success possible.

Christine and I look forward to meeting with many of you during the quarter. And we'll be planning at least one road show to accomplish this objective. If you're interested in meeting with us, please contact our IR Council, Deborah Stapleton, at deb@stapleton.com.

Thanks again, and goodbye.

And this concludes today's conference call. Once again, on behalf of CEVA, thank you all for your participation. You may disconnect at this time.