MoSys, Inc.
Q3 2013 Earnings Call Transcript

Published: 18 Oct 2013

[Operator:]

Good morning, and welcome to the MoSys Third Quarter 2013 Financial Results Conference Call. [Operator Instructions] As a reminder, this conference is being recorded today, Friday, October 18, 2013. I would now like to turn the call over to Beverly Twing of Shelton Group, Investor Relations. Please go ahead.

[Beverly Twing:]

Thank you, Lacey. Joining me today on today's call are Len Perham, MoSys' President and Chief Executive Officer; and Jim Sullivan, Chief Financial Officer. Before we begin today's call, I would like to remind everyone that this conference call will contain forward-looking statements based on certain assumptions and expectations of future events that are subject to risks and uncertainties. Such statements are made in reliance upon the Safe Harbor provisions of Section 27A of the Securities Act of 1933 and Section 21E of the Securities Exchange Act of 1934, which include, but are not limited to, benefits and performance expected from use of the company's embedded memory and interface technologies and ICs; expectations concerning the company's execution and results; expected benefits of the company's ICs; product development; achievement of design wins and timing of shipments of the company's ICs; predictions concerning the growth of the company's business and future markets; and business prospects, strategies, objectives, expectations or beliefs. Forward-looking statements made during this call are subject to risks and uncertainties that could cause actual results to differ materially from those projected. Additional information concerning factors that could cause actual results to differ materially from any forward-looking statements made during this call are contained in the company's most recent reports on Form 10-K and Form 10-Q filed with the Securities and Exchange Act -- Exchange Commission, in particular, in the section titled Risk Factors and in other reports that the company files from time to time with the Securities and Exchange Commission. MoSys undertakes no obligation to publicly update any forward-looking statement for any reason, except as required by law, even as new information becomes available or other events occur in the future. Thank you for your attention. I will now turn the call over to Len Perham, Chief Executive Officer of MoSys. Please go ahead, Len.

[Leonard Charles Perham:]

Thank you, Bev. Good morning, everyone. I'll begin today's call with a few highlights and then provide more detail on our third quarter activities. Jim will then review our third quarter financial results prior to opening the call for your questions. And I'll probably have a couple of closing remarks. Outstanding operational execution, that is, bringing out new products that work first time right out of the box, coupled with increased amount -- or a couple of the increased amount of face time we are getting with our customers now that we have multiple product lines to discuss with them, leads me to feel very good about our overall progress up to and through the third fiscal quarter of 2013. Other than the fact that I cannot be certain how fast our customers' revenue will ramp up once they release their new products to the market, I am pleased with the progress we've made throughout this last reporting period. From period to period, our production roadmap becomes more robust, which allows us to not only pursue many new design win opportunities for our existing products, but to discuss in substantial details the requirements of our prospective customers' future networking systems. We are gratified to see our Bandwidth Engine 1 design win starting to ramp into production and anticipate a reasonable increase in units shipped in the fourth quarter over third. Currently, among other important goals, we are intensely focused on operational execution; keeping new product development on schedule; system bring-up, that is, helping our existing customers ramp their end users up into production and get going down the road; and cost control. All in all, it has been a productive quarter, and I'm satisfied, quite satisfied, with our progress. A little about design wins. Bandwidth Engine 2 continues to outpace Bandwidth Engine 1 in the sales funnel as customers more opt -- more and more opt for the MSR620 Burst mode device, MSR720 access mode device or MSR820 intelligence-based device, all products which are derived from our base BE2 architecture. Additionally, as reported last quarter, once a customer has adopted the Bandwidth Engine solution and comes to understand the uniqueness of the GCI interface, the customer is very likely to use it again and again. To this end, we are gratified to add an additional design, a third platform win from our Tier 1 customer during the quarter being reported here today, and as well, we can report yet another design win out of Asia as well. At this point, I am reasonably certain that we will exceed 10 Bandwidth Engine design wins for calendar year 2013, probably in the range of 10 to 15, something like that, and that's discounting the wins from 2012. Turning now to product updates. Bandwidth Engine 2. We've now completed both the functional and performance level characterization in BE2, and we remain confident that we will pass the final quality assurance and reliability benchmark testing in the fourth quarter of 2013. We remain on schedule to do a full production release of BE2 before year end 2013. In addition to having all 3 Burst -- all 3 products, that is, a Burst mode, access mode and intelligence mode configurations of BE2, we will also offer the part in 3 I/O speed grades

[James W. Sullivan:]

Thank you, Len, and good morning, everyone. During the course of my comments, I will make several references to non-GAAP measures. Unless otherwise indicated, each reference will be an -- to an amount that excludes stock-based compensation expense and intangible asset amortization. These non-GAAP financial measures and a reconciliation of the differences between them and comparable GAAP measures are presented in our press release and related current report on Form 8-K, which was filed with the Securities and Exchange Commission this morning and can be found at the Investor Relations section of our website. Now let's review our third quarter financial results. Total revenue was $1 million compared with $1.1 million for the second quarter of 2013 and $1.3 million for the third quarter of 2012. Licensing and other revenues for the third quarter was approximately $0.2 million, consistent with both the previous quarter and the third quarter of 2012. As we've mentioned on previous calls, licensing revenue is comprised of revenue recognition and prior IP agreements as we complete final deliveries and provide related maintenance and support services. We expect this revenue to continue to taper off and be offset by increasing revenue from our IC products, which is also included in licensing and other revenue. The third quarter of 2013 was the first quarter in which IC revenues comprised the majority of licensing and other revenue, and we expect this trend to continue. We remain focused on further growing our revenue attributable to shipments of our Bandwidth Engine and LineSpeed integrated circuits through product development and sales efforts. Royalty revenue for the third quarter of 2013 was $0.8 million compared with $1 million for the previous quarter and $1.1 million for the third quarter of 2012. The sequential decrease in royalty revenue was due primarily to a decrease in shipments from a foundry licensee. Similar to licensing revenue, royalty revenue is derived from our legacy IP business. However, we expect to continue receiving royalties on these projects even though visibility to this future revenue stream is limited. Third quarter 2013 royalty revenue was recognized from 14 licensees compared with 13 in the second quarter of 2013. GAAP gross margin for the third quarter of 2013 was 83% compared with 93% for the prior quarter and 96% a year ago. The sequential decrease in gross margin was primarily due to higher onetime manufacturing costs associated with our IC products. In terms of our operating expenses for the third quarter, research and development expenses were $6.2 million, compared with $6 million in the previous quarter and $7 million in the third quarter of 2012. The sequential increase in R&D expenses reflected higher costs related to the tape-out of another version of our initial Gearbox products and higher professional service expenses, partially offset by lower Bandwidth Engine 2 back-end costs and CAD tool expenses. Selling, general and administrative expenses were $1.6 million, compared with $1.5 million in the previous quarter and $1.7 million in the year-ago period. The sequential increase in SG&A expenses was primarily attributable to higher stock-based compensation expense. Total operating expenses on a GAAP basis for the third quarter of 2013 was $7.8 million and included $0.3 million for amortization of intangible assets and $1 million in stock-based compensation expense. This compared with $7.4 million in the previous quarter and $7.3 million in the third quarter of 2012. On a non-GAAP basis, total operating expenses for the third quarter of 2013 were $6.6 million, compared with $6.2 million in both the second quarter of 2013 and the year-ago period. On a GAAP basis, net loss for the third quarter of 2013 was $6.9 million, or $0.14 per share, compared with a net loss of $6.4 million, or $0.15 per share, in the prior quarter and a net loss of $6.1 million, or $0.15 per share, for the third quarter of 2012. On a non-GAAP basis, net loss for the third quarter was $5.7 million, or $0.12 per share, which excludes intangible assets, amortization and stock-based compensation expenses totaling $1.3 million, compared with a non-GAAP net loss of $5.2 million, or $0.12 per share, in the previous quarter and a loss of $4.9 million, or $0.12 per share, in the year-ago period. Net loss per share for the third quarter of 2013 on a GAAP and non-GAAP basis was computed using approximately 48.2 million weighted average shares outstanding. Now turning to the balance sheet. As of September 30, 2013, our cash and investments balance was $53.3 million, compared with $58.9 million at June 30, 2013. Cash burn in the third quarter approximated $5.7 million and included approximately $0.6 million in tape-out costs related to our LineSpeed product and $0.2 million in payments related to our May 2013 equity offering. As of September 30, our total headcount was 104 employees, compared with 97 for the previous quarter. Of our total employee count, more than 80% are in engineering and research and development, and 24 are located in India, compared with 17 in the previous quarter. This concludes my prepared remarks. At this time, we would like to open the call for a question-and-answer session. Operator?

[Operator:]

[Operator Instructions] And our first question will come from the line of Gray (sic) [Gary] Mobley with Benchmark.

[Gary W. Mobley:]

This is Gary, actually. I had a couple of questions relating to your comments in your press release regarding a, I think, "meaningful" unit ramp for Bandwidth Engine. Just hoping maybe you can -- hoping we'd delve into that a little bit more. And first clarify with me whether or not the majority of your licensing and other revenue was integrated circuits-related sales? And then, again, how meaningful might the ramp be in the fourth quarter compared to the third quarter?

[James W. Sullivan:]

Yes. In the -- Gary, in the licensing and other, the majority was integrated circuit revenue. And at this stage, as I've made comments, I think, in the Q&A in previous calls, we do still have a fairly high revenue reserve given the early stage of shipments, et cetera, and given where we're at in these shipments. But we do expect that, as I said on the call, to continue. At this point, we do expect a meaningful increase in units. But putting that in perspective, we're still obviously in the early stages and awaiting full production from our customers. But to put it in some numbers, we're talking about, at least from the orders we have in the backlog right now, a doubling of units shipped quarter-over-quarter.

[Leonard Charles Perham:]

And let me add to that, Gary, I made the comment, this -- basically for the first time today, we discarded calendar year or fiscal year '12 design wins, and we talked about what was going to happen in fiscal '13. And we even made a comment -- well, I'm going to make a comment in closing about fiscal '14 as well. But we also said in our transcript today or -- that we still have a ways to go to -- before our customers are strongly ramping into production. And we don't want to mislead our investors that we're going to be cash flow positive next quarter. That won't happen. But it isn't unreasonable that Jim is going to be right, that period over period now, we're going to see substantial, on a percentage basis, increases in units shipped. And we're entering into a whole new phase now of the rebirth of MoSys as a fabless semiconductor company. That's -- we waited a while, and it's now beginning to happen. So that's one...

[Gary W. Mobley:]

Okay, all right. Could you provide some distinction between the average selling price for Bandwidth Engine 1 and Bandwidth Engine 2? And have those average selling prices for each respective product changed all that much? And have you changed your expectations regarding the gross margin potential for the products? And then the last question for me, Len, is can you discuss competitive landscape out there, whether it be competition relative to RLDRAM 3 or Memory Cube? Or could you talk about what you're generally competing with when vying for these design wins?

[Leonard Charles Perham:]

Okay. So basically, for budgetary purposes in-house, we would typically say that Bandwidth Engine 1 is going to be selling in the range of $80 to $90, and we would say that Bandwidth Engine 2 is probably going to be selling in the range of $110 to $130. Bandwidth Engine 3, which is a ways out yet, will sell for substantially more than that. And that's a larger memory, and it's got an enormous amount of intelligence built into it. But -- and those prices haven't changed too much. Of course, we're shipping limited -- well, let me just put it another way. We're shipping probably above both of those prices today, but we're still at low volume compared to what we expect in the future. So the prices I just gave you are budgetary prices for large orders, and so that's a little bit on the pricing. A guy's going to get into trouble if he tries to figure out what we're doing with BE at $80 and figuring BE2 at $110 to $120, something like that. And your third question was what again? Just remind me.

[James W. Sullivan:]

Competitive landscape.

[Leonard Charles Perham:]

Oh, competitive -- I'm all -- sorry, Gary -- competitive landscape. So basically, in the competitive landscape area there's still people trying to do it by using enormous number of pins and do what now? They're trying to solve their problem set for header processing with larger and larger arrays of RLDRAMs or in many cases, maybe even multichip modules or some in-house. On the part of our network equipment company, many are even using some in-house solution that they developed years ago when the -- something like the Bandwidth Engine didn't exist. Down in the packet processing area, forgetting the header processing now, but down in the data packet area, we see that, that is an area that doesn't require such high-speed access. And down there, we compete on occasion with the Hybrid Memory Cube. We haven't seen much from this high-bandwidth memory, but the Hybrid Memory Cube has generated some interest. It's -- it has the same problems that any DRAM solution has. It has some latency issues and so on and so forth. And of course, this stack of memories called the Hybrid Memory Cube has a lot of technological innovation in a way of through-silicon vias and so on, and that's probably unproven and may cost a fair bit of money. But we do run into multichip module solutions down in -- down where access isn't as important as sheer capacity of memory. And my final comment will be, we haven't seen anyone come out with a monolithic solution yet that offers the same access speed and bandwidth preservation that our -- or effective use of bandwidth, if you will, as our Bandwidth Engine family. And mostly, what comes at us now is either just continuing to do it the old way or trying to solve some of the problems with multichip module solutions. And we think we're a smaller footprint and perhaps lower power and a better cost basis for our customers than those alternatives. Perhaps that's an answer.

[Operator:]

[Operator Instructions] And our next question will come from the line of Jeff Schreiner with Feltl and Company.

[Jeffrey A. Schreiner:]

Yes. Len, how many customers are now signed up with Bandwidth Engine? And could you give us a breakdown in terms of the number of customers between Bandwidth Engine 1 and Bandwidth Engine 2?

[Leonard Charles Perham:]

First off, I think we have something in the order of -- when I say that we're going to do something in the order of 10 to 15 design wins for the year, I believe that we're at something in the area of around 9 different customers right now, something like that. We may not always report that, but at this point, I think that's not unreasonable. So that's probably around 9, Jeff, something in that area. And the second part of your question was?

[Jeffrey A. Schreiner:]

Just what's the breakdown of -- within those 9 customers of BE1 and BE2?

[Leonard Charles Perham:]

Okay. In the end of -- in fiscal 2012, we talked about something, I think, in the order of 5 platform wins, 5 design wins. And those are the ones that are entering early production ramps now. So we would expect that through 2014, the majority of our shipments would be of Bandwidth Engine 1 -- and then it would probably cross over in 2015. In terms of what's on our backlog, by sheer volume of units, recognize that it's still not massive yet, but by sheer volume of units, it'll probably be BE1. But in terms of number of customers we ship to, I think probably I'm going to guess that half of the customers that we'll ship to between now and year end will be using BE2 and the other guys that are earlier in their ramps. So -- and if we look into the sales funnel, there's -- because of the fact that BE2 allows you to bring off an access, Burst or macro-based device, customers are much more likely to target exactly what they need now. If they want to deal with all the subscription, they'll take a Burst mode device. If they're talking about header processing, they'll go right to a access-based device. So I think in the future, new orders that aren't booked today that we haven't called design wins are probably now moving up to 60% to 70% BE2, something like that, Jeff.

[Jeffrey A. Schreiner:]

Okay. And then how many Tier 1 customers -- out of the customer base you just discussed, Len, how many of those will be Tier 1 customers for MoSys? Or how many Tier 1 customers do you believe, based on your visibility, you may have over the next 12 months?

[Leonard Charles Perham:]

Over the next 12 months, I think that we're going to have -- do business with the -- we'll do -- we'll have -- we'll be calling more than half of and perhaps -- I'll say -- I'll be conservative and say maybe not all of, but maybe more than half of the Tier 1 guys will become customers in 2014. And we have to recognize, of course, it'll take some roadmap for their new systems to go into production, and they'll ramp, in some cases, a little further out. But in our short-term pipeline, even this year, we may yet see another Bandwidth Engine -- another Tier 1 customer come onboard.

[Jeffrey A. Schreiner:]

Okay. Then final question for me, Len. I just wanted to talk -- get your thoughts that -- I'm thinking that maybe part of your answer was targeting this, but trying to understand a product like LLDRAM who's worked -- who's being used by a very large Tier 1 networking company. I was wondering how LLDRAM competes against Bandwidth Engine.

[Leonard Charles Perham:]

LLDRAM is the same thing as RLDRAM. I think RLDRAM is probably a Micron product and LLDRAM is a Renesas product. There are -- I don't -- I think that's the right way around. And in my opinion, based on all my years of experience, it's a -- it's possibly a -- it's a DRAM solution, and it's possibly a 2-transistor solution instead of where a DRAM would be a single transistor -- these things might now be 2 transistors per cell in an effort to get the performance out of the DRAM that's needed for header processing. They have some latency issues, and they have some access issues. So basically, it's my opinion that the -- where the QDR memory is probably, I'm guessing, now steeply declining in today's systems and looking forward, they're -- I think that the LLDRAM or RLDRAM is really struggling. I believe that we position BE1 against RLDRAM Gen2, and BE2 was sort of positioned against RLDRAM Gen3. And in terms of our Bandwidth Engine 3, which is capable of something in the area of 8 billion to 10 billion accesses per second, there hasn't been any response from the RLDRAM or the LLDRAM camp, and I think that's because there is none. So I think we're -- I think that's been a long standard, and I think it's getting fairly long in the tooth that the performance levels we're looking at are for future equipment.

[Operator:]

And our next question will come from the line of Krishna Shankar with Roth Capital.

[Krishna Shankar:]

Yes. Len and Jim, congratulations on the design win momentum and a good engineering execution. A couple of questions. You mentioned 10 to 15 potential customers for Bandwidth Engine this year. I'm presuming most of that is Bandwidth Engine 2. Can you talk about the eventual sort of revenue opportunity over the next 3 to 5 years with these 10 to 15 customers? What could be the potential revenue opportunity assuming sort of a typical production ramp in next-generation networking systems?

[Leonard Charles Perham:]

So first off, Krishna, you -- I made the comment that our 10 to 15 design wins this year would be predominantly all Bandwidth Engine, but I didn't make the statement they would be absolutely all Bandwidth Engine, so I want to be very clear. The -- we're going to do 10 to 15 design wins this year, new design wins, and we couple them all together, and we don't separate them out. And since BE -- since the LineSpeed products are very new, you can assume that at this point in time, it's predominantly Bandwidth Engine. But I want to be straight up and say that, in 2014, if we should have a significant increase, which I do expect we will, then a fair number of those new ones are going to be LineSpeed products. Now going back to your revenue, we don't provide any guidance on that yet, and it would be -- and I made the comment on the -- either in the press release or in the discussion this morning that it's a little bit difficult for us to see the full production ramp of our customers. In many cases now, we're working directly with the customers and their customers as they ramp these big boxes into production. We're everything but out at the end user site, if you will, and it's very exciting, and it's showing us how good our part is. So we don't give -- we're not in the position to give a lot of guidance on these numbers. I don't know if you want to comment on that or not, Jim.

[James W. Sullivan:]

Yes, I would agree with your position on that, Len. We're not done on the ability to -- right now to kind of give that outlook in guidance. I would just highlight the point Len made in the script and in the press release in that once we're -- once the customer has chosen to opt for, in the case of looking at Bandwidth Engine, a serial memory solution, made the decision to implement it, the opportunities just grow -- compound as far as the ability for us to have additional wins with the customer and the ability for us to gain those wins with, in some cases, even minimal effort from our sales and marketing and applications team because basically, we're in there. We're kind of on the shelf and available for use by the teams. And I think, as Len highlighted, we saw another one of those design wins in the quarter. And that's the part that's hard to gauge because some of these things are happening and, in some cases, it's pleasing because our guys don't even have to spend a lot of time on it. They know -- the customer knows the parts and the capabilities and is looking at it for other applications. I mean, the other thing that's always hard, too, is with these design wins, the volume estimates are -- we get estimates from the customers, but we really don't know how accurate they are. And the ones that you think might be small could be big and vice versa. And at this point, it's really a numbers game as just far -- as far as gaining as many as we can and breaking into new customers. So we do have that opportunity for the follow-on wins.

[Leonard Charles Perham:]

Krishna, I would add to that, that we're talking about populating line cards with a packet processing memory and packet -- header processing memory and packet buffering memory and oversubscription and queuing and statistics. And so probably these design wins are something between 5 and even 10 or more of our products per board. And these things sell anywhere from $25,000 to $250,000 each, these boards. It's a multi-, multi-, multibillion dollar industry, billion line cards for these advanced systems. And if you -- we have mentioned on other calls that if you take a look at the model for a NetLogic or a Cavium or even the early days of IDT when it was just breaking into the specialty memory business many years ago, you can see how the ramp might be. The unknown is how fast these big box guys release into production. But I can tell you that things should be pretty exciting around here by the end of 2014 and going into '15 compared to what's nonoperationally today.

[Krishna Shankar:]

Great, that's helpful. And then on the LineSpeed product line, you indicated good traction both in terms of design wins and customers moving into evaluation and qualification. Those, I guess, to some extent, target either existing sockets or existing applications where you don't need a complete set of platform redesigns. So can you give us some sense for what types of applications near term you might see success with the LineSpeed Gearbox? And what the slope of that ramp might be?

[Leonard Charles Perham:]

Again, the LineSpeed products are sitting in the speed path of the line card itself. We've mentioned oftentimes that the key thing about the Gearbox is the ability of the part to work into some significant load in terms of loss, and we've measured in decibels, and that very short reach is 12-decibel loss. And short reach would be 15, and long reach is, I think, it's 30. And we were very gratified that our first product offering was capable of both short reach and very long reach. On the other hand, the second critical thing, besides being able to operate into a significant load, is how low can you get the power to be. And of course, if you're trying to work into a heavier and heavier load, then you're burning more and more power. So some of the guys that have focused right on short reach, very short reach, have lower power than we do. We might have briefly mentioned today that -- and I've mentioned in other forums and so forth, that we would expect to come out with our second-generation LineSpeed products perhaps in the second quarter of '14. These -- the first round of products was to see if we had the capability of being in the business, and we're extremely excited about the success we've had. And the second generation of products starting in the first or second quarter of next year, we intend to be a major player. There are -- these sockets are served by various customers today, very good and powerful companies, Infi, Broadcom among those. We don't expect to dominate it, but we expect to take our fair share. I think the question you're asking is, are we competing for existing sockets? And the answer is yes. It could ramp up slightly faster, but still, we're always going to have the problem of how fast these big boxes go into production. We're very, very gratified. I didn't get very specific about it, but at this European Conference on Optical Communication, there were 9 of these interoperability tests done, where somebody would drive into our product that where our card embodying our product is. And then it would feed out, and in many cases, we went to 10 kilometers of fiber to the receiving end. And we were -- there were 9 of these interoperability tests done, and we were on 5 of them. And we were a -- and we had significance -- significantly excellent performance of our Gearbox and Retimer. It was -- it really made us look very, very good, and it's going to generate a ton of excitement. I think the guys just got home the beginning of this current week, actually. So perhaps that's an update on what you were asking.

[Operator:]

At this time, I show that we have no further questions in queue. I would now like to turn the call back to Len Perham for closing comments.

[Leonard Charles Perham:]

So first off, I want to thank you for all -- all of you for calling in today and finding what's -- finding out what's going on over here. And I'm just going to make a couple of comments, and it's a bit referencing back to all the things we've said to each other this morning. I said that 2013 should see us win 10 or more new design wins, predominantly with Bandwidth Engine. The LineSpeed is just starting. It's a significant percentage of what's going on in the sales funnel as things move down to become design wins. And we should expect a very exciting '14 for Bandwidth Engine, and then it will be added to in terms of additional wins coming from the new family of LineSpeed products. 2014 should see us as -- and I probably said this, too, should us as perhaps as much as double the number of design wins. It's a very exciting prospect indeed. We're entering a whole new phase of MoSys rebirth as a fabless semiconductor company. And we may get frustrated with each other at the ramp of growth somewhere along the way, but I can tell you that things have changed, and it's going be -- it's a different world now. I want to thank you very, very much for taking the time to be with us this morning. And we look forward to speaking with you guys again in 3 months, and we really appreciate your support. And again, thanks for dialing in and letting us tell you our message. Have a great day, everyone.

[Operator:]

Thank you for your participation in today's conference. This concludes the presentation. You may all disconnect. Good day, everyone.