MACOM Technology Solutions Holdings Inc (MTSI) 2003 Q4 法說會逐字稿

The following is a recording of the Raouf Halim teleconference with Mindspeed for Wednesday, October 29th, year 2003 at 4:00pm central time.

Excuse me, everyone. I would like to introduce Simon Biddiscombe, Chief Financial Officer of Mindspeed, who will chair this afternoon's conference call.

Please be aware that your line is in a listen-only mode.

At the conclusion of Mr. Biddiscombe's presentation, we will open the floor for questions. At that time, instructions will be given as the procedure to follow if you would like to ask a question.

I would now like to turn the conference over to Mr. Biddiscombe. Sir, you may begin.

Thank you, Moon (ph) .

I would like to welcome everyone to our fourth quarter conference call, discussing the results for our first full quarter as a separate public company.

Joining me on the call today is Raouf Halim, our Chief Executive Officer.

I will begin our call with a review of our quarterly income statement and balance sheet. Raouf will then provide a product and market overview and our outlook for the current quarter. We will then open the call for your questions.

Before I begin, I want to remind you that our comments today will include statements relating to future results that are forward-looking statements as defined in the Private Securities Litigation Reform Act of 1995. Actual results may differ materially from those projected as a result of certain risks and uncertainties, including, but not limited to those noted in our earnings release and other various filings with the SEC.

I would also like to remind everyone that the results we will discuss today are from the pro forma income statement before amortization of intangible assets, special charges, uncertain other non-operating gains and losses. We believe this approach provides the best insight into the operating performance of our company. A reconciliation of the historical pro forma information, to GAAP, is included in our earnings release, a copy of which is available on our website at www.mindspeed.com.

Turning now to our financial results for the fourth fiscal quarter. Mindspeed delivered significantly improved financial performance over the prior quarter in all key financial areas. Today we announced fourth quarter revenues of $23.2 million -- up 15% sequentially from the prior quarter, and at the high end of the guidance range we provided one quarter ago. This sequential revenue growth was driven by increased demand across the board for our four key product families. Sales of our ATM/MPLS network processors solutions and T/E carrier transmission products increased sequentially in line with our expectations, and reflecting the continued recovery in our traditional wire line markets, both in the U.S. and overseas.

Our multi-service access voice processors and high performance analog product families experienced significant double-digit revenue growth, as these two product franchises continue to capture share in the emerging high-growth voiceover IP and fiber-to-the-home markets.

Raouf will provide more color on revenues in his comments.

Gross margin was $16.3 million, or 70% of revenues, including a roughly four percentage point benefit from the sale of products written off in fiscal 2001. This performance represented the sixth consecutive quarter in which we've delivered gross margins in excess of our long-term model of 65% percent. The gross margin exceeded our expectations, primarily as a result of favorable product mix.

Operating expenses of $34.5 million reflect a decrease of $4.2 million, or 11% from expenses of $38.7 million in the prior quarter. We exceed our expectations of at least a three million dollar reduction, primarily by completing some of our previously announced cost reduction initiatives ahead of schedule.

We remain on track to achieve our target quarterly operating expense goal of $31 million exiting this calendar year, and thereby realize the full benefit of the cost reduction initiatives in the March, 2004 quarter.

Our pro forma operating loss was $18.2 million -- an improvement of 27% over the prior quarter's operating loss of $25 million, and well ahead of the expectations that we set at the beginning of the quarter.

Other expenses and a provision for foreign income taxes and the aggregate resulted in a net credit of $259,000. We would expect these items to have a minor impact on the income statement for the foreseeable future. As a result, our pro forma net loss improved 27% percent over the prior quarter to $17.9 million, or 20 cents per share, based on approximately 91.6 million shares outstanding.

Turning now to the balance sheet. Our total cash consumption for the September quarter was $21.3 million. This represents a 34% reduction from last quarter's cash consumption of $32.4 million. Last quarter's cash consumption excludes cash received from Conexant Systems during that period, including cash received at the time of separation. Cash consumption from operations was $25.1 million, including approximately two million dollars in cash outflows associated with becoming a separate public company, such as insurance payments, and $3.9 million in restructuring payments associated with our previously announced cost reduction initiatives. Net cash generated by financing activities was $3.7 million, principally consisting of $6.1 million in proceeds from the exercise of employee stock options, offset by amounts repaid to Conexant Systems of $2.1 million in accordance with the terms of our separation agreement. Our cash and cash equivalents at the end of the quarter totaled $80.1 million.

Capital expenditures were less than $100,000, and depreciation was $3.2 million. We would expect quarterly capital expenditures to average between one and two million dollars per quarter moving forward.

In terms of working capital, receivables were $11.7 million, resulting in DSOs of 45 days, in line with our long-range target and representing a significant improvement from last quarter's performance of 54 days. We reduced net inventories by one million dollars in the quarter to four million dollars, resulting in inventory turns of 68, exceeding our target range of five to six turns. Gross inventory, including amounts previously written off, totaled approximately $68 million at the end of the quarter.

Restructuring reserves exiting the quarter were $10.8 million. Cash payments associated with these reserves are primarily related to EDA tools and facilities that are no longer in use. We're in active negotiations to significantly reduce cash outflows associated with these obligations, and in fact, have successfully renegotiated a contract with one of our primary EDA tool providers.

I would now like to turn the call over to Raouf for his comments on the quarter.

Thank you, Simon.

I'm very pleased with our performance in our first quarter as a public company. Every one of our four key product families grew sequentially, and we continue to make excellent progress in executing on our growth strategy towards the return to profitability, as well as encapturing share in our target metro/access market applications.

Our results this past quarter further reinforce our belief that the telecom carrier markets are in the early stages of a sustainable recovery worldwide. Our customers are increasingly optimistic as they execute on a large number of significant new platform developments where Mindspeed products have been designed in.

During the quarter, we experienced continued new product revenue growth from market-leading customers worldwide. We further diversified our revenue base from both an application as well as a geographical perspective, and we secured a large number of very significant tier-one design wins across our broad product portfolio. More specifically, new products launched in fiscal year 2001 and thereafter contributed more than 75% of our sequential revenue growth in Q4. We experienced a particularly strong pick-up in demand for voice-over IP, high performance analog applications, such as fiber-to-the-home, and DS3/E3 carrier solutions.

From a product mix perspective, revenues from these markets contributed the majority of the growth this past quarter, and we believe that they will continue a disproportionate share of our revenue growth going forward.

Design wins grew dramatically over Q3, with half of our design wins coming from key Asia/Pacific customers. On a dollar-weighted basis, nearly half of our design wins this quarter and throughout the past fiscal year were outside of our traditional wire-line markets in new applications, such as storage, fiber-to-the-home, and wireless infrastructure.

From a geographic perspective, we are particularly pleased that revenues in the Asia/Pacific region grew sequentially in excess of 30% for the second quarter in a row. Asia/Pacific revenues are now on par with the Americas at 43% each, with Europe making up the balance.

Key customers driving revenues this past quarter included Cisco Systems, Nortel, McDATA, Lucent, Demons (ph) , Alcatel, Huawei Technologies, and multiple Asian ODMs of optical modules supplying key OEMs, including Hitachi, Fujitsu, and NEC, Cisco was again our only ten percent customer, moving both direct and indirect sales through third parties.

I would now like to review results for each of our four key product families before providing guidance, starting with our high-performance analog product portfolio.

Revenues grew substantially, driven by shipments of physical media dependent, or PMD devices, including laser drivers and pre- and post-amplifiers, supporting the ongoing deployment of fiber-to-the-home services in Japan, and the build-out of metropolitan networks in China. Key PMD customers this past quarter included OEMs, such as Huawei Technologies, Fujitsu and Sumitomo, as well as optical module manufacturers, such as WTD in China, Delta in Taiwan, and many others.

Fiber-to-the-home is the next generation broadband market opportunity. It includes both point to point and point to multi-point or passive optical PON networks. After many years of technology development and cost reductions, fiber-to-the-home is now transitioning to mainstream; broadband subscriber deployments worldwide.

The Asia-Pacific region leads fibers to the home deployments with 64% of global subscribers, according to a recent report from RHK, an independent market research firm. The government broadband initiative in Japan is targeting 10 million subscribers by 2005 from less than 600,000 today. NTT is reportedly adding 80,000 subscribers a month, and accelerating to reach over a million subscribers per year.

South Korea has a similar initiative, with a goal of five million subscribers by 2007. And China, meanwhile, has seven ongoing PON trials in major cities, and Taiwan is likely to deploy fiber-to-the-home as well.

Even in the U.S., prospects for fiber-based broadband networks appear more promising with the recent regulatory industry initiatives.

Mindspeed has captured a significant share of the fiber-to-the-home market with their broad portfolio of PMD products, which have already been widely deployed in carrier networks, such as NTT in Japan.

Turning to enterprise applications of high-performance analog products, storage area networking remains a key market for our high-performance cross-point switches, and we believe we are ideally positioned to benefit, as the industry transitions from two-gigabit to four-gigabit per second fiber channel solutions.

During the past quarter, we broadened our industry-leading family of cross-point switches with the introduction of new two-by-two and four-by-four devices, offering high-speed switching capabilities across a wider range of local as well as wide area networking applications. These new analog cross-point switches integrate our patented amplified signal conditioning technology and include integrated multi-rate CDR circuits. These new devices enable our customers to use a single solution for multiple market applications.

Further strengthening our high-performance analog market position, we captured multiple high-value analog product design wins this past quarter for applications including fiber-to-the-home, storage area networking and access aggregations at key OEMs such as Sienna, Venisar (ph) , Fujitsu, and Samsung, Inrange -- now CMT -- Lucent, as well as NEC, amongst others.

Now, turning to our second product family. Revenues from multi-service access products grew dramatically as we ramped shipments of voice processors to customers such as Siemens, Huawei Technologies and DT in China for both telecom carrier as well as enterprise applications, particularly for deployments in the Asia-Pacific region in countries specifically, such as Japan and China.

We also remain on track to commence the early ramp of our Chagall IP PBX on a chip late this calendar year, leading to production in the first half of calendar 2004. We believe voice-over IP is a compelling technology that will eventually replace many of the circuit switches in phone networks around the world. This market represents a significant long-term opportunity for Mindspeed, given that today voice-over IP accounts for less than 3% of global voice phone calls, according to industry estimates. Voice-over IP is also growing rapidly in the enterprise markets now that it has reached total quality, and businesses take advantage of the tremendous cost and productivity gains offered by voice-over IP.

Worldwide spending on IP phone systems by businesses is expected to more than double this year to one billion dollars, capturing nearly 20% of the total enterprise phone markets, according to the independent market research firm, Delaura (ph) Group.

Mindspeed is uniquely positioned in this market and we are experiencing accelerating design activity this past quarter for our high-density voice-over IP Miro processor and wireless base station controller, as well as wire line carrier gateways, including wins at tier-one carrier equipment OEMs, such as Siemens, GTE, LGE and many others.

We also secured multiple voice-over IP design wins with Huawei Technologies in the past quarter. In addition, we scored multiple design wins with Chagall for enterprise equipment, including voice-over IP access gateways, IADs and PBXs at key customers, such as Harbor Networks in China, DTE, Telrad in Israel and many others.

Turning to our T/E carrier portfolio. Revenues grew sequentially, and in particular, revenues from our DS3/E3 products nearly doubled. We experienced strong demand from the deployment of multi-service provisioning platforms, or MSPPs, as well as optical edge devices, or OEDs, as carriers continue to upgrade and extend their networks. Specifically, our DS3/E3 LIUs are shipping in virtually all of the DS3 line cards for next generation metro equipment, for example, including Nortel's OPTera Metro and Lucent's Metropolis DMX platforms, for deployment and carrier networks, including AT&T, Bell South and Qwest.

During the quarter, we further our strengthened our T/E carrier market position with a large number of significant design wins in metro carrier, edge aggregation, and wireless applications.

As linecard densities begin to grow, customers are increasingly attracted to our high-port count, highly-integrated DS3/E3 LIU and integrated LIU framer solutions, and we believe we are winning a disproportionate share of new DS3/E3 design opportunities.

In Q4, we further increased our share of the DS3/E3 market with significant new design wins at market leaders, including Cisco Systems, Nortel, Alcatel, Juniper, LGE, DTE, Tellabs, and many others. Additionally, we won multiple other T1/E1 and HDLC designs with OEMs, such as, UT Starcom, Harbor Networks in China, Huawei Technologies, Nortel, Fujitsu, LGE, as well as Lucent.

Finally, in our ATM/MPLS processor portfolio, revenues grew sequentially for the second quarter in a row. We experienced strong demand from key customers, including Cisco Systems, Nortel, Alcatel, Lucent, and Fiber Home in China, among a number of others for use in their access router, DSLAM as well as wireless infrastructure equipment.

Two weeks ago, we announced the third generation of our highly-successful traffic stream processor family. Offering the broadest range of software compatible network processing solutions in the industry, our TSP3 family consists of four distinct traffic management and layer-two internet working processors, with speeds ranging from 155 megabits or OC3, to 2.5 gigabits, or OC48, and the suite of Ethernet-enabled firmware applications. These new products will enable the fine grain quality of service to allow carriers to provision new services across a full range of packet and cell-based networks.

The TSP3 offers our equipment customers the ability to use one system's design across multiple port speeds for a wide range of applications in enterprise, access, as well as metropolitan networks. This protects their engineering investment and allows them to quickly enhance, as well as add new features, with downloadable software upgrades, while enabling us to significantly increase the size and scope of the markets we address with our TSP solutions.

We believe we are gaining share in the market for ATM/MPLS processing, and during the past quarter, we captured key design wins for carrier infrastructure, as well as enterprise LAN equipment, with many OEMs, including LGE, Juniper Networks, as well as multiple design wins at Huawei Technologies.

In conclusion, I'm very pleased with our performance across the board this past quarter, and in particular, the execution on our growth strategy. I'm even more confident than ever in our growth prospects, given the continuing improvement in the fundamentals underlying our core markets, as well as our continued market share gains across every one of our four key product families.

As I mentioned at the outset, we believe that we are in the early stages of a sustainable recovery in the telecom carrier markets worldwide for the past two years. And order trends, which were strong throughout the summer, have carried through into the current quarter. In addition, our design win pipeline is improving, with a large number of significant new customer platform developments underway in which Mindspeed products have been designed in.

We expect the revenue growth drivers for Mindspeed going forward to be voice-over IP, as well as our high-performance analog products addressing such applications as fiber-to-the-home, and storage area networking.

These product cycles are just starting and will be complimented by the ongoing recovery of our T/E carrier and ATM product franchises as we continue to capture share in high-growth metro/access markets, particularly for DS3/E3 carrier applications.

Based on all of these spots of indicators, we remain quite confident that we will achieve our pro forma operating profit breakeven levels of approximately $45 million in quarterly revenues before the end of calendar year 2004.

And now turning to our expectations for the current quarter. We again expect our first quarter revenues to be up 10 to 15% sequentially. We also expect overall gross margins to be roughly 68%, and we anticipate further lowering our total operating expenses by roughly two million dollars this quarter as we continue to make progress in our cost reduction initiatives. As a result, we expect to improve our pro forma operating loss by 15 to 20%, and expect to further reduce our cash consumption by at least 25% this quarter.

That concludes our formal comments today. Operator, let's please open the line for questions.

At this time, we will open the floor for questions. If you would like to ask a question, please press the star key, followed by the one key on your touchtone phone now. Questions will be taken in the order in which they are received. If at any time you would like to remove yourself from the question queue, press star, two.

Our first question comes from Matt Suess (ph) with Credit Suisse First Boston Corporation.

Hi. It's Jeff Waspur (ph) for Matt.

Can you talk about your linearity; how order patterns track throughout the quarter, and whether you saw a pickup in September?

Yeah. This is Raouf. The linearity was fairly consistent with prior quarters; nothing unusual to report. I think in general it's fair to say that we did not experience quite the decline in August that has been typical in past years. August was little more robust than we would have expected, and September was quite strong. Nothing dramatic to report, you know, in terms of changes in linearity. Overall, it's quite consistent. It was a good quarter.

OK. And you didn't mention this in the commentary, but what are you guys seeing in the DSL business?

The DSL business remains quite strong, but in particular in Asia-Pacific, with deployments -- significant deployments ongoing in China, a little bit of a leveling off in Korea, but huge growth in China. We're designed into a large number of DLAMS, with many of the same products I talked about at some of the leading providers of DSLAMS into those countries. And so, we are certainly seeing strong deployments there as well.

And then, finally, when you gave your December quarter guidance, how much of that is seasonality versus inventory buying versus real demand?

Jeff (ph) , this is Simon.

I think the reality is that most of the demand is real demand at this point in time. We haven't seen, we don't believe, any significant change in inventory in the channel with our ultimate customers. So, I don't thing there's anything that we're seeing that suggesting the growth is coming from inventory replenishment at this point in time. Our best perspective on what's going on is that our customers are buying for consumption at this point in time.

OK. Thanks a lot.

Our next question comes from Mark Grossman with Needham & Company.

Great. Thanks.

It sounds like you guys were looking for the MSA business to grow fairly rapidly next year, but it sounds like it's happening a lot sooner than you were expecting. Can you talk about the timing of when that started to pick up? And then, for the Miro and Chagall products, what do you think the ramp up of those products look like over the next few quarters?

OK. So, Mark, this is Raouf.

We have been anticipating for quite some time now that the voice-over IP market will start ramping. We have certainly been very engaged in the marketplace now for a long time, frankly, and we have garnered a significant number of tier-one design wins.

It's always very difficult to predict the exact ramp, if you will, and the timing of volume deployments for our customers. It is difficult for them as well; given the lack of visibility they often encounter with field trials and so forth from service providers.

We're quite pleased that in fact in this past quarter we saw a significant amount of volume of our Miro carrier class voice-over IP products, as well as older generations, or prior generations of voice-over IP processors. Both turn on. And, you know, I think it's fair to say that we believe this momentum will continue in the current quarter and extend beyond that. We have captured, as I said, a large number of tier-one design wins and a significant market share, particularly where it counts, which is in Asia-Pacific.

All right. OK. And then, in the lower-speed optical components for fiber-to-the-home, can you talk about what the competitive landscape looks there now?

Yeah, certainly. So, the segment that we're focused on -- service starts at 10/100. They're fast Ethernet rates, and pops out at roughly OC12 or 622 megabit rates, with each pod (ph) being around fast Ethernet OC3. And in that segment, those exactly are the speeds that carriers are deploying today for fiber-to-the-prem and fiber-to-the-home deployments. So that, if you will, is the sweet spot of the marketplace.

The competitive environment is quite simple to characterize for you. It's predominantly ourselves, Maxim and then, to a much lesser extent, a couple of smaller players in this space. But we see Maxim just about everywhere we go. So, we think we're head to head with Maxim in that space.

OK. Great. On the storage side for switching, you've had one big customer. What do you think the prospects are for expanding to get other customers in that market?

Mark, in Q4 we scored another significant tier-one design win with cross points, which is specifically in the storage space. Unfortunately, we're not at liberty to disclose the name of that OEM that we scored with, but we're quite pleased with that progress. And that's coupled with a significant amount of traction with our fiber channel TMD products as well in module applications, so we continue to make inroads in the storage space.

OK. Great. And last one. I guess you don't give out a book to bill, but could you give us some sense for how much current orders you need to make the guidance, or how well the backlog is?

Yeah. The one comment I'll make, Mark, and then I'll turn it over to Simon quickly, but certainly our backlog at this point in the game is higher than the backlog that we had for the same week last quarter. But beyond that, I'll turn it over to Simon to give you a little more color on that.

What I was gonna add to Raouf has just said, Mark, is that, you know, we're not gonna give book to bills, but suffice it to say that all of the key metrics around revenues are moving in the right direction relative to where they were last quarter, and that's what gives us the degree of comfort over being able to give the 10 to 15% guidance that we've given at this point in time.

OK. Thanks a lot, guys.

Our next question comes from Jeremy Bunting with Thomas Weisel Partners.

Hi, guys. Thanks. It's Herman Roy (ph) for Jeremy.

I had a follow-up on the MSA business. Raouf, you mentioned a few times that the strength is coming out of Asia, and I was wondering if the number of data points are indicating now that there's some VOP strength in other geographies. Any comment on that? Give us some more color on where some future strength going into '04 could come from, and if some of that strength is built into your growth expectations for the business. Thanks.

Certainly. Yes, to date I think it's fair to say that in our opinion, most of the voice-over IP deployments have been in Asia-Pacific. Outside of A-Pac historically it has been quite limited, and primarily in field trials, or limited to low--volume field trials types of deployments. And again, I'm talking about the carrier space specifically.

Going into '04, to answer your question, we think there is gonna be some real deployments, volume deployments, of next-gen voice-over IP in North America. Certainly, a lot of very positive indicators in that regard, with tenders starting to go down by carriers such as Verizon, as I'm sure you're aware, as well as others in North America. Those have not yet percolated into real volume shipments at this time, but we expect that '04 will be the year that we will see voice-over IP become reality in North America. So, we're increasingly bullish about the carrier space on a global basis.

I would add to that that in the enterprise segment, it is a global phenomenon already. There's certainly clear evidence of converged voice data networking in small office, branch office installations today in the U.S., as well as virtually every part of the world. And we expect that there's going to be a continued transition to carrier-grade IP PBXs, either from the ground up new platforms that are IP architectures, or retro-fits to existing TBM PBXs to support independent IP subscribers.

Simon, you mentioned 4% benefit to gross margins from the written-off inventory. Can you remind us of how much longer you're gonna have the benefit coming out of written-off inventory? Thanks.

Unfortunately, that's a bit like predicting how long a string is. It could certainly go on way out there into the future. We have, as I mentioned, around $68 million of gross inventory at this point in time, including all the stuff that's being written down to zero. So, we could enjoy that benefit for a significant period of time.

Got it. And the head count at the end of the quarter?

The head count at the quarter was 650.

Thanks very much, guys.

Our next question comes from Carter Driscoll with Independent Research Group.

Hi, guys. Congratulations for a very solid quarter.

Do you have any good flush out on the fiber-to-the-premises argument, and kind of the timing differences between deployments, specifically in Japan and North America, and when you think it might become a North American story, and really its effects on other access technologies, and will it displace them or be complimentary?

Yes, certainly. So, I think we laid out some of the statistics around our expectations of voice-over IP installations. But, Carter, just starting at the top, I think it's clearly the case now that the incremental cost per subscriber of deploying low-end optical -- meaning sort of fast Ethernet 10/100 or even OC355 megabits -- through the premises, whether that's an apartment or a dwelling of some kind or a small business -- clearly the economics of doing that have become a lot more interesting, a lot more, I guess I would say compelling, relative to other technologies over the past -- just literally the past 12 months or so. And we are seeing that in a number of countries the alternative of deploying low-end fiber as an access technology is not only being taken seriously, but is actually being adopted as an alternative to other technologies that may be cable-based, like DSL as a for instance.

And we are clearly seeing that countries, such as Japan, have adopted fiber-to-the-prem as a key access technology. And, you know, there's a lot of public information floating out there from carriers, such as NTT specifically, regarding their plans to deploy fiber-to-the-prem -- an increasing mix of fiber-to-the-prem and fiber-to-the-home relative to DSL within the broadband subscriber base.

So, we think it's a combination of cost, as well as the compelling services that can be deployed over that kind of a bandwidth, that make it pretty attractive.

And it to date, you know, it has been primarily Japan, as you alluded to. We are seeing now a number of major cities in China -- I would give you, for instance, the city of Shanghai -- where there are very large field trials that I think in many books would qualify, if you will, as real deployments of fiber-to-the-prem actually ongoing right now. And, as I mentioned in my comments, we see a large number of countries -- Korea being another notable example -- where fiber-based access is viewed as the next generation broadband carrier topology.

Do we think that fiber is happening? We know fiber is happening. So far, it's been primarily limited to the countries that I rattled off in Asia. We see a lot of progress here in North America. You may be aware of some of the recent rulings and certainly the Joint Purchasing Consortium, the JPC, that have come together to put together a common set of requirements around fiber-to-the-prem deployments, access deployments in the United States, which we think is very encouraging and is somewhat reminiscent of the early days of DSL going back six or seven years ago.

And so, we think, you know, the stage is set for fiber, in particular PON, or passive optical networks, to really become the next generation broadband access technology of choice on a global basis.

And specifically, though, to answer your last question about North America, I mean, we remain somewhat -- I don't know what the word is or the term is -- maybe cautiously optimistic. But, I mean, I think it's gonna happen, Carter, but I think, you know, there's still a lot of work that the ILECs have to put into this business model, including, you know, the business side of the arrangements for the content delivery, the tarriffing (ph) , the servicing, et cetera. It is not just about the access medium. There's a lot more that has to get wrapped around this business model for it to be successful in North America.

So, we think they've got a lot of work ahead of them, beyond just laying the fiber and installing the optoelectronics.

So, it might be a year to two years before it really happens in North America. We're certainly not counting on a big turn-on of fiber-to-the-home in North America, you know, anytime in the next four to six quarters, but we have a very, very significant market share overseas in PMDs particularly, that, you know, ranges in many cases, you know, north of 50%, and in some cases 80% market share in these optical modules. We see this as a big positive for our business.

Do you see the adoption rate from when you get qualified to ramp as similar to other types of products, you know, somewhere between 12 and 18 months?

No. It is actually considerably shorter than that. The primary customer segment consists of module ODMs, original design manufacturers, of optical modules, as I mentioned, both in China as well as in Taiwan.

Um-hmm.

Their design cycle is typically half of what you mentioned -- sort of six to nine month kind of development cycle, and then their shipping.

We have been qualified by I think virtually every meaningful ODM manufacturer of optical modules for fiber-to-the-home deployments. We are now deployed, quite widely deployed, within NTT's network, which, as you may be aware, has very, very stringent requirements.

Um-hmm.

So, we're considered, in many cases, to be the de facto standard for the electronics in those markets.

So, you could actually expect that there might be some type of ramp in the second half of '04 that we might not have built in yet, specifically because they're already being trialed? Or, is it fit to be potentially pushed in the fourth quarter of calendar '04, or '05?

Well, actually, we are deploying and we have been deploying for a couple of quarters now; but, again, into Asia-Pacific. I'm not sure if your question was specifically around North America.

No. More really about Asia.

Oh, Asia is happening right now, Carter.

No, I understand it's happening, but I'm saying a much more robust ramp kind of like in the first half timeframe, or second half, or -- you know, I guess I'm trying to zero in on what could be a really significant out-performer on your model.

Well, we expect that the markets will continue to grow throughout calendar year '04. I think there is a potential in the second half of '04, as carriers such as, you know, carriers in China and in particular, Korea, such as Korea Telecom and a couple of other guys, as they qualify this technology and get a high level of confidence with it to actually turn on. So far, it's been limited to Japan.

Yup.

And so, I think, yes, in the second half of '04 there is a potential for somewhat of an increase in the deployments of fiber-to-the-home, particularly in Asia. And then, as we said earlier, maybe in '05, '06 you'll start to see this phenomena turn on in the United States.

Great. Thanks, gentlemen.

Our next question comes from Charlie Glavin with ThinkEquity Partners.

Hey, guys.

Hey, Raouf, in terms of some of the deployments out of Asia, without sounding too greedy, but it seems as if there's been several unannounced designs with some fairly large guys -- not necessarily yourself, but others. It doesn't sound like there's an inventory issue, but do you get a sense that some of the Asian OEMs are trying to get through this last calendar quarter and flushing out some of the older products before they announce new platforms, particularly within China, and also shipments out of China into Japan?

That's a pretty insightful question, Charlie. I mean, I really can't say that there's any level of meaningful inventory of older products right now in almost any part of Asia, to be honest with you. And speaking for the particular product technologies that we address -- I can't talk to other technologies very well; I'm talking specifically to our portfolio.

So, it's hard to say, you know, that they're holding back just on the account of inventory or trying to move what they've got out of the way. I think we've seen design activity pick up, you know, consistently throughout this calendar year. I don't think there's anything limiting or holding back OEMs, such as Huawei and ZT in China, as a for instance, or Alcatel, Shanghai Bell, or even in Korea from guys like Samsung and Hyundai LGE, et cetera. There's nothing keeping them back other than just their own limitations on getting products, you know, developed, tested and out the door at this point in the game.

Typically, it hasn't been that big a -- you know, that high level of fanfare of equipment announcements out of Asia as there has been out of, say, North America and occasionally Europe. I'm not sure, though, that I would expect quite the level of visibility as we may have been accustomed to in the past.

OK. Shifting -- within the North America, and down ever so slightly, but in talking with you, even though Asia was 43%, any guess in terms of whether the consumption is now closer to 50 to 60%, or whether or not Asia is actually, beginning to accelerate. When we had talked previously, you had said -- back when Asia was about 38% of your revenue in terms of point of shipment, it was actually closer to 50%? Any variance on that since the last quarter or earlier this year?

Well, certainly with the growth that we have experienced, particularly in our two product franchises -- fiber-to-the-home and voice-over IP, which, at this point in the game, are heavily Asia-Pacific oriented, as we talked about. We think that while our own mix was, as I said, 43%, end shipments by both Asian OEMs, as well as OEMs here in the U.S., are probably representative of north of 50% of the end consumption. I'd be very hard-pressed to give you a number. I mean, a wild flag might be something around 60%, but I just don't know, Charlie.

OK.

But, it's got to be higher than what it was previously.

Charlie, what I would add to that is that is if your question was has it changed significantly throughout the quarter, just the last quarter; yeah, there probably has been a shift in the business pool of Asia from a consumption perspective over the course of the last quarter, and that is the case.

Yeah, and that seems to be consistent with what we're hearing.

One last question, Raouf. In terms of the voice-over IP within North America, when we talked with you, along with Avaya and Catapult, one of the things that came up was not only possibly some of the transitions within Culsiners (ph) , but also the emergence of security and security issues over necessarily just QOS (ph) . Has there been any significant shift that you've seen, and in particular, given some of the hybrid solutions out there, are you looking for your IP PBX on a chip as being key for the ramp within North America, either in late '04 or '05 for a true voice-over IP, not the hybrids?

Yeah, absolutely. The short answer to your question is yes, absolutely. I mean embedded security within an overall voice and data solution is absolutely a hard market requirement now. We think we're ahead of the curve with both our Chagall products, as well as follow-on products that embed that capability. And, yes, we think it's critical to our market share capture in North America, and we think we're doing real well with that at this point in the game.

I'm sorry, Raouf. I was asking more in terms of -- obviously, you're well situated -- but in terms of the industry positioning, in order for your products to take traction.

I see. We think it's in the -- yes, I think in terms of full-on ramp of full IP networks and full IP infrastructure to the enterprise, yes, those problems have got to be solved first, before carriers and particularly Fortune 50 (ph) companies, are comfortable installing a full IP PBX and all its terminals -- that is not the current market situation. I think most large corporations are deploying IP capability incrementally to PBM PBXs today. But, you're absolutely right that encryption and the security types of issues have got to be solved for, you know, the large corporations to deploy full IP solutions. And I think we're on the way there.

Great. Thanks, Raouf. Thanks, Simon.

Mr. Biddiscombe, there are no more questions at this time.

Thank you, operator.

That concludes our conference call today. On behalf of all of us at Mindspeed, thank you for participating this afternoon. We look forward to updating you on our performance next quarter. Good bye.

This concludes today's conference.