萊迪思半導體 (LSCC) 2015 Q4 法說會逐字稿

Good day.

My name is Victoria and I will be your conference operator.

At this time, I would like to welcome everyone to the Lattice Semiconductor fourth quarter 2015 conference call.

(Operator Instructions)

After the call has ended, if you would like to listen to the replay of this conference, you may access it by dialing 404-537-3406 and use conference ID 41056653.

Thank you.

I would now like to turn the call over to David Pasquale of Global IR Partners.

Sir, you may begin.

Thank you, Operator.

Welcome, everyone, to Lattice Semiconductor's fourth quarter and full-year 2015 results conference call.

Joining us today from the Company are Mr. Darin Billerbeck, Lattice's President and CEO, and Mr. Joe Bedewi, Lattice's Chief Financial Officer.

Both executives will be available for Q&A after the prepared comments.

If you have not yet received a copy of today's results release, please email Global IR Partners, using LSCC@GlobalIRPartners.com, or you can get a copy of the release off of the Investor Relations section of Lattice Semiconductor's website.

Please note, we also just published a PowerPoint presentation on the IR site to accompany today's call.

The slides will be referenced in management's prepared comments.

If you could now turn to slide 2, please, which covers the Safe Harbor statement.

Before we begin the formal remarks, I will review the Safe Harbor statement.

It is our intention that this call will comply with the requirements of SEC Regulation FD.

This call includes and constitutes the Company's official guidance for the first quarter and full-year 2016.

If at any time after this call we communicate any material changes to this guidance, we intend that such updates will be done using a public forum, such as a press release or publicly announced conference call.

The matters that we discuss today, other than historical information, include forward-looking statements relating to our future financial performance and other performance expectations.

Investors are cautioned that forward-looking statements are neither promises nor guarantees.

They involve risks and uncertainties that may cause actual results to differ materially from those projected in the forward-looking statements.

Some of those risks and uncertainties are detailed in our filings with the Securities and Exchange Commission, including our Form 10-K for the fiscal year ended January 30, 2015 and our quarterly reports on Form 10-Q.

The Company disclaims any obligation to publicly update or revise any such forward-looking statements to reflect events or circumstances that occur after this call.

Our prepared remarks also will be presented within the requirements of SEC Regulation G regarding Generally Accepted Accounting Principles, or GAAP.

Some financial information presented by us during the call will be provided on both a GAAP and a non-GAAP basis.

By disclosing certain non-GAAP information, management intends to provide investors with additional information to permit further analysis of the Company's performance for results and underlying trends.

Management uses non-GAAP measures to better assess the operating performance and to establish operational goals.

Non-GAAP information should not be viewed by investors as a substitute for data prepared in accordance with GAAP.

If we use any non-GAAP financial measures during the call, you will find that the required presentation of and reconciliation to the most directly comparable GAAP financial measure is in the Company's earnings press release.

I would like to now turn the call over to Mr. Darin Billerbeck.

Please go ahead, sir.

Thank you, David, and thanks for everyone for joining us on our call today.

I'm going to start my comments on page 3 of the slides.

Results for the fourth quarter came in as expected, but we were surprised by a payment delay relating to an IP sale which occurred during the quarter.

Because of this delay, we were unable to recognize $7 million of forecasted revenue, resulting in our revenue and non-GAAP gross margin being below our expectations.

We expect to recognize the revenue and associated cost of sales when future payments are received.

Overall, Q4 GAAP revenue came in at $101.2 million, which was down 7.8% from Q3.

Our Q4 non-GAAP gross margins was 54.6%, compared to 55.7%.

Our non-GAAP net loss was $0.04 per share in both Q4 and Q3.

GAAP revenue was $406 million for the full year 2015, an increase of 10.9% over the full year of 2014.

Revenue was higher due to the addition of the Silicon Image products and licensing revenue as a result of the acquisition in early March of 2015.

Our non-GAAP gross margin for the full year was 56.2% and our non-GAAP net loss was $0.13 per share.

2015 ended up with double-digit growth from 2014, as we mentioned previously, but down from our original estimates when we acquired Silicon Image.

On the consumer side, revenue dropped significantly for both our FPGA and imaging products at Samsung, as our customers struggle with profitability and began focusing newer phone models at the lower end of the market.

This lower cost focused approach reduced the higher end feature sets that we provide.

In the communications market, Huawei and ZTE combined dropped almost 25% in FPGA shipments, as LTE buildouts in China softened.

Industrial for the year was our highlight, increasing on a non-GAAP basis almost 7%, due to the strength in Asia.

As bad as 2015 was, we delivered on our aggressive synergy goals.

We will benefit from the savings in 2016 of the $49 million in synergy savings actioned in 2015, along with our 13% headcount reduction.

We clearly understand that we need to have both a growth strategy while remaining focused on the cost structure.

We have been and are continuing to take actions along the way to ensure we can hit our operating income goal of 20%.

We expect to hit that goal in the back end of 2016.

If you turn to slide 4, I will give you some additional color on where we are at with respect to the overall acquisition.

Earlier in 2015, we announced the acquisition of Silicon Image.

This was a great fit for Lattice from a market, product and customer perspective.

On the surface, many people questioned combining an ASSP company with a programmable logic company.

If you believe video as being the killer application, the benefits are clear.

Video is data and FPGAs are the best hardware acceleration on the planet for data.

Fast data?

Just look at Intel's recent acquisition of Altera.

Now that we have the video expertise, a platform to accelerate, and both wired and wireless ways to distribute it, we have a greater reach into markets didn't exist for us in the past.

After nearly one year into this deal, we're even more confident in the fit.

If you turn to slide 5, you can see a visual of how many numerous -- how numerous our actual deal opportunities are.

These are the markets where Lattice has competitive advantages, based on our proven solutions and existing customer traction.

We're a global semiconductor company, but more than that, we help manufacturers made products that are smarter, connect faster and better.

We help your cell phone recognize your voice and then download a movie in seconds without wires.

We help you stream that movie from your cell phone to the world's biggest ultra high-definition TV and then easily connect that TV to your home theater system.

In the factory, we make the machines easier for you to interface with, give them higher definition eyes, and help them network together.

When you send an e-mail from your smartphone, we allow the cell phone towers to be smaller, consume less power, and connect to data centers faster.

In the data center that routes your e-mail, we help you monitor the health of the servers.

So while you don't see our products, you're interacting with them all day, every day.

We make your experience smarter and more connected.

Turning to slide 6, we can touch on the consumer market.

In consumer, we provide solutions for smartphones, tablets, Z TVs, wearable, and even new spaces, like drones.

Did you know that we're designed into one of the largest drone manufacturers?

Not only do we handle the multiple interfaces and control, they're exploring our millimeter wave solutions and wireless camera technologies.

Why have wires in your wireless drone?

Consumer is in our DNA.

This is a rapidly changing market where the time to market is key.

The flexibility, programmability and cost savings afforded by our programmable and imaging solutions make Lattice an ideal partner.

Some examples of where we win and what we do.

A close collaboration with Qualcomm enabled us to be part of their quick charge reference design for USB 3.1, a strategic move that opens the door to be part of ODMs and OEMs worldwide.

Design wins in the latest voice-activated speakers to LED phone covers to ear sets to AVR next generation 4K, and even in the most popular credit card readers, we're in all of them.

Our customers are driving to deliver richer and more responsive experiences, ranging from higher resolution video to more intelligence in their devices.

Take the latest MediaTek reference design, where Lattice can be found delivering 4K30 on USB 3.0 today, heading towards 4K60 in USB 3.1 tomorrow.

Our ultra low power small phone factory FPGAs are the perfect solutions.

Our millimeter wave devices, including side beam snap, provide wireless connection technologies and can transfer high def content to mobile devices in seconds.

Now let's not forget about the HDTV business.

From a video streaming to the latest HDMI standard, to sound bars and AVR equipment, Lattice is the leader in transmitting the latest and greatest standard to any device, whether it moves or it's mounted on your wall.

By working with Lattice, our customers get a design advantage[s] and can take products to market faster than ever before.

If you please turn to page 7, you can see we are winning in communications and computing.

Customers need to connect anything to everything at even higher data rate.

Networks require progressively higher bandwidth and increased reliability with the transition to cloud-based infrastructure.

Lattice's programmable solutions are optimized for I/O expansion, acceleration and hardware management.

In the server space, we've won the next generation rate controller large OEM with the MachXO3 over the other guys.

In addition, one of the fast-growing server companies in China has chosen Lattice to provide enclosure management and control functions.

In the communications market, our newest ECP5 product line is gaining ground in the next generation of backhaul, wireless and headnet solutions, where our low-power small form factor and affordability fits well with big ASICs, small FPGA architectures.

Add to that our millimeter wave solutions and we now can have discussions with the same communications customers about wireless point-to-point solutions for the backhaul, along with what they need for 5G.

We are excited about the revolutionary beam stirring technology.

We are making point-to-point links lighter, cheaper, lower power and easier to install.

Think of this as enabling backhaul at wireless fiber data rates.

If you please turn to page 8. We see many opportunities in the industrial and automotive markets, increasing amounts of data required to be gathered, connected, processed and displayed.

We form the backbone of several integrated solutions, including complete HD camera, DVR and human machine interface solutions, digital TV [sign-ins], surveillance and medical displays.

Our performance tested and regulatory approved YHD modules can eliminate the clutter and unreliability of wires.

This is critical for low latency applications where Y Gig doesn't fit.

Examples of the latest wins.

We serve a bridge solution for the latest Segway Mini Pro, which debuted during Intel's keynote at CES.

We expect to begin shipping our YHD solution in the Lenovo detachable micro projector beginning this year, also.

We're designed into navigation systems, rear view mirror replacements and wireless transmission systems for sterile medical environments.

And our automotive qualified MHL HDMI video connectivity ASPs complement our FPGAs, allowing customers to stream ultra high-definition video from their mobile phones to their in -are entertainment systems for the ultimate connected car experience.

So you can see why we're so optimistic about the business and really excited about the diversity of the opportunities in front of us.

Please turn to Slide 9. As we anticipated during the acquisition of Silicon Image, there are many places were FPGAs and image products can coexist to create more stickiness.

Some examples.

In the consumer space, virtual reality headsets need to convert from HDMI to red green blue back to MIPI, and then have a right and left display.

You could do that with a combination of imaging and FPGA solutions.

4G base live TV boxes that need to convert back from an application processor to MIPI to red green blue back to HDMI for your high-def TV, display, tablet or smartphone, another FPGA plus image combo solutions.

In the industrial space, digital signage has multiple new HDMI inputs that need to be converted back to earlier HDMI specs and then convert back to LEDS.

Everywhere there's an image, there's an interface with FPGAs convert to just about any standard.

Again, a combo solution.

And finally, in developing multi element antennas in the communications space, the easiest way to accomplish this is using good old [serties] based FPGAs.

We have that, too.

Slide 10, please.

We continue to execute and deliver on innovative solutions to the market needs, next generation eyes products that are always on and always listening for wearables, smartphone and tablets.

More capability packed into even smaller form factors and even lower power.

The newest IP built on older product lines, like MachXO2 and XO3, to support mobile and industrial video standards.

Faster versions of new products like our five gigabits per second 30s ECP5 products, not only good fits for big ASIC and small FPGAs in the communications market, but also used for video image bridging in new and existing video interfaces.

Our latest and greatest millimeter wave products that can inter operate with Qualcomm's Y Gig.

Yes, we have that, too, along with proximity of wireless USB 3.0 that can synch and transfer high-def content in seconds.

Our strategic plan is simple, grow in our markets, achieve the revenue synergies, and execute on our 20% operating income goal.

Let me now turn the call over to Joe for details on our financial results.

Joe?

Thanks, Darin.

If you turn to Slide 11, I'm going to touch on the financial highlights from the quarter.

As part of our press release, we have provided detailed reconciliations of GAAP to non-GAAP financial measures.

For the fiscal fourth quarter of 2015, revenue was $101.2 million on a GAAP basis and $101.3 million on a non-GAAP basis.

Gross margin for fiscal Q4 2015 was 53.5% on a GAAP basis and 54.6% on a non-GAAP basis.

When compared to Q3 GAAP gross margin of 54.5%, the gross margin benefit we achieved from improved mix and the sell through of previously reserved products was more than offset by increased period costs, the effects of lower manufacturing volume in Q3, and Q3 scrap recovery which did not occur in Q4.

The increase in period costs includes costs associated with ramping dual source fab and back-end capability, which we expect to incur through Q2 2016.

Total non-GAAP operating expenses for the fourth quarter were $51.9 million, excluding $3.5 million in restructuring charges, $0.4 million in acquisition-related expenses, $8.8 million in amortization of acquired intangible assets, $21.7 million in impairment of goodwill and intangible assets, and $4.4 million in stock-based compensation.

I'd like to provide a bit more information related to the impairment charge we incurred in Q4.

During the quarter, there were two separate impairment triggers associated with the Qterics long-lived assets.

First, the Company made the decision to no longer pursue the data services business.

In line with the process we followed related to capital allocation and operational effectiveness, we made the decision that resources would be better utilized in our core business operations.

Second, the Company agreed to redeem an equity investment in Qterics made by Qualcomm.

Following our assessment of impairment of long-lived assets of Qterics, we determined that Qterics long-lived assets had been impaired as of January 2, 2016 and that this impairment was other than temporary.

As a result, we recorded an impairment charge of $21.7 million, which is comprised of $12.7 million of goodwill, $3.9 million in developed technology, and customer relationships of $5.1 million, to write these assets to their estimated fair market value.

Income tax expense for the quarter was $3.5 million.

Cash tax expense was approximately $2.9 million on a GAAP basis.

Reflecting the aggregate impact of the various items mentioned, we recorded a net loss for the fourth quarter of approximately $45.5 million, or a loss of $0.38 per basic and diluted share.

On a non-GAAP basis, our net loss was $5 million, or $0.04 per basic and diluted share.

For the quarter, basic and diluted share count was approximately 118.1 million shares, net cash used in operating activities was $2.3 million in Q4.

We ended the quarter with cash and investments of approximately $102.6 million.

Accounts receivable increased to $88.5 million at the end of Q4, as compared to $85.3 million at the end of Q3.

Days sales outstanding were 80 days, compared to 71 days last quarter.

Inventory at quarter-end was $75.9 million, compared to $79 million at the end of Q3.

Months of inventory stands at 4.8 months, compared to 4.8 months at the end of Q3.

We spent approximately $6.6 million on capital expenditures and incurred $17.9 million in depreciation and amortization expense during the quarter, compared to $4.6 million and $17.6 million, respectively, in Q3.

Interest expense for the quarter was $5.5 million.

Turning to slide 12, you can see some of our financial highlights for the year.

In addition to what we covered earlier, this shows the non-GAAP OpEx breakout.

We believe we are in a good position entering 2016, given the continued headwinds we expect to face in the first half of the year.

We believe we are at a bottoming of revenue and that our operating structure is right sized for this environment.

Importantly, we think we have considerable leverage, in that we can support higher revenue levels without increasing overhead substantially.

Our focus remains on optimizing our operational spending at a lower stabilized revenue run rate to generate appropriate free cash flow and enhance shareholder value.

This concludes the financial review portion of the call.

I'm going to turn it back over to Darin for the outlook on slide 13.

Darin?

Thank you, Joe.

In terms of our specific expectations on a non-GAAP basis, revenue for the first quarter of 2016 is expected to be between approximately $95 million and $101 million on a non-GAAP basis, with revenue for the full year 2016 expected to be approximately $460 million to $470 million on a non-GAAP basis.

Non-GAAP gross margin percentage for both the first quarter and the full year of 2016 is expected to be approximately 56%, plus or minus 2%.

Total operating expenses, including acquisition- or restructuring-related charges, are expected to be approximately $49.6 million, plus or minus 3% on a non-GAAP basis for the first quarter of 2016, and approximately $175 million, plus or minus 3% on a non-GAAP basis for the full year 2016, which includes the benefit of synergy savings.

Restructuring charges are expected to be approximately $2.1 million for the first quarter of 2016 and approximately $2.8 million for the full year 2016.

Acquisition-related charges, including the amortization of acquired intangible assets, are expected to be approximately $8.9 million in the first quarter of 2016 and approximately $35.8 million for the full year 2016.

In summary, the positive impact of our major transformational moves in 2015 have positioned us for long-term success.

We expect Q1 to represent a low point for 2016, and we are well positioned for growth in the second half of 2016, led by existing design wins in our three key markets, consumer, communications and industrial.

If the macro improves, we should do even better.

We will be well positioned to address the numerous and diverse opportunities we outlined in the call today with our expanded solutions and our competitive advantages.

We're focused on the ongoing execution of our business strategy and remain fully committed to driving higher returns for all of our shareholders, which I continue to be the largest individual shareholder.

That concludes our prepared remarks.

Operator, we will now be happy to take any questions.

(Operator Instructions)

Tristan Gerra, Baird.

Good afternoon.

Your full-year revenue guidance implies a 20% type of sequential growth for the next couple of quarters in the [role] later this year.

What's the basis for the assumption that your second half is going to be so strong relative to prior years, and also what type of year-over-year growth are you expecting in your consumer business this year?

Between 2015 and 2016, Tristan, it's going to be a lot of consumer.

There is some industrial in there also.

So we're going to consider consumer to be up, we're going to consider industrial to be up, communications is probably going to be down, we're going to call that 5% to 10%, and then IP is roughly flat.

Okay.

And is the delayed IP payment embedded in your Q1 revenue guidance?

No.

Not materially.

Okay.

And then finally, could you give us an update on your 1180 wireless technology in terms of potential timing and when it gets material to the top line?

Yes, I think the nice thing is Qualcomm and Intel seem to be working together, which is going to help the MC ecosystem, because Qualcomm has been leading with the AD solutions.

And as you are aware, we are very interoperability, we have a lot of interoperability between Qualcomm.

So if Intel and Qualcomm end up with the same [Mac Fi] and all the software, then we're in good position.

We announced at CES a dongle for Qualcomm's Y Gig which works today.

We worked with their factory in Israel to make sure that we worked functionally and from a data transfer perspective.

So we expect to roll out that dongle as the demand comes in.

Again, Qualcomm is going to be integrating that on mobile devices.

Intel has already announced the integration of their millimeter wave solutions on the laptop front.

So there will be plenty of devices to connect to after that.

Great.

Thank you.

Richard Shannon, Craig-Hallum.

Good afternoon, Darin and Joe.

Thank you for taking my questions.

I'll follow up on one of the last questions here, thinking about your growth profile for the year, particularly in the second half.

My guess is talking about consumer first as part of the drivers for the sequence of growth throughout the rest of the year.

Is it safe to assume that will be the big part of the growth driver, or should we see more balance between that and industrial, any more color you can give us on that, please?

Yes, consumer will be a bigger driver.

So obviously, as Joe had mentioned, there's going to be some gross margin pressure because of that stuff, but we're not considering it, again, you've got a good mix of consumer, industrial and some of the [IPE] to offset that.

So that will actually end up for a nice year for us.

But yes, consumer will be the big trigger, but let's not count out industrial and some of the growth there.

Okay.

I want to step away from the revenue part of that question, just ask another quick one on gross margin.

So a decent part of your consumer related product lines, at least historically, have been below corporate average gross margins.

And your guidance for the year is the same for the first quarter.

Can you help us resolve that?

Can you help us figure out or understand why that would not be under a little bit of pressure as you grow consumer in the second half?

There is some pressure there for sure.

But you still have the IP and royalty volume that's going to come through, as well as industrial uptick.

And we have some product, I will call it evolutions, that will help us on the margin side, also.

Okay.

Fair enough.

Back again to the consumer bucket, maybe delving into that a little bit more.

You've got a number of different pieces here, both from the Lattice, as well as Silicon Image side.

Curious, any one of these going to be bigger driver than others?

I can see how your eyes product line could be a big driver, potentially maybe even some larger customer wins, but sounds like you're getting a little more excited about Y Gig, and then where does USB [text e] fit into that?

So let's start with the consumer.

Most of the consumer growth will be FPGAs this year.

So it will be consumer FPGAs, which is something we've been working on for a while to reestablish, post the Samsung era.

And then you've got the -- we do still see a lot, although not as large as Samsung, a lot of the MHL in some of the emerging countries, the reference design that I alluded to with MediaTek, there's an MHL reference design out there also that ships into India, where they do use their smart phones as productivity enhancements for them, because they don't have laptops and PCs.

The USB Type C, we're still seeing a lot of video over USB 3.0 out of the micro, some of the micro USBs, not necessarily as big of a USB 3.1.

However, to talk out of the other side of my mouth, Microsoft did announce a whole slew of families that use USB 3.1, and we're in those.

So the key for us is really the adoption of that through multiple different platforms being the big guys adopting 3.1 and then being able to establish ourselves as the keys to running video out of that platform.

Because a lot of people don't necessarily want to burden the phones with the video in the phone, but they want to be able to convert it once it comes out, whereas tablets and some of the other higher end want to have the embedded features of being able to establish the video as a video output out of those.

So we're going to see an HDMI-MHL consumer front, which is really an emerging market, and maybe tablets and those things, but the big driver for consumer is going to be FPGAs being ICE.

Okay.

Perfect.

Maybe a couple of quick questions and I will jump online.

I think you mentioned some costs that you're bearing out throughout this year as you go to some dual source manufacturing.

Can you help us understand what's going on there?

What's driving this?

Is this more of an internal push, or is this something that's being requested by some customers here?

That's what we started talking about in Q3, where we were moving to dual source with fab and back end capability.

It was driven by customers.

And we are playing it out; basically it will be completed by the second half -- or the second quarter of this year.

Second quarter of this year.

Okay.

Fair enough.

And my last quick question, Joe, what amount of cash flow do you expect to use to pay down debt here?

What are your parameters?

How are you thinking about that?

What kind of cash flow do we expect to see?

Yes, how much will you pay down debt this year, do you expect?

Based on this model, we have the 75% free cash flow sweep.

We won't have any obligation to pay debt based on last year, because of the cash flow being virtually negative because of the consolidation.

Going into 2016, the $460 million to $470 million model, what you see if you look at the Street numbers, we're going to spend somewhere in the $50 million range on cash is what we are anticipating free cash flow.

75% of that would go to pay down debt.

Okay.

Perfect.

I think that's all my questions.

I will jump in line.

Thank you.

[Pardee Ho], Everbright.

Hello, Darin and Joe.

Thanks for taking my call.

I have a question on the IOT and the virtual reality market.

So would you mind commenting on them?

And also how would the ICE product be used in these devices?

And then my next question would be, how would your recovery of the TXMC facilities affecting your product schedule?

So let's start with the last question first.

The TXMC doesn't affect us at all.

There was a very small percentage of scrap post the earthquake.

So we won't see any impacts, and a lot of the parts that we had were past some of the steps there, so I think we will be okay there, minimal disruption.

If you look at the virtual reality headsets, they typically are combination.

What you will see is an FPGA may use as an I/O expansion, which means the input will come in and it would be spent for your left and right eye.

So that's a lot of times what you will see from an FPGA.

The HDMI image will actually be processed and then modified using a SIM G product.

So you'll see both the SIM G product in those virtual headsets, plus an FPGA as the I/O expansion and the left and right splitter.

So we have combo solutions for that.

They're not going to ship in super high volume, but they do prove that in these particular applications, it's a great way for us to get some stickiness with the combination of the two products.

And then additionally, there's also some wireless versions of that using YHD that you can put into some of those virtual reality headsets, much more expensive, but then you are not stuck with a wire sitting on your virtual reality headset.

So we're working on that, both on 10ADP versions and then we're also working on the 4K versions of that.

So that will be a much more expensive solution, but it may be sought after because of the limitations of the wire on your virtual reality headsets.

Okay.

Good.

Thank you.

And also, you had talked about the Type-C a little bit earlier.

So I would like to understand more about how much do you expect that would be contributing to the bottom line, and what you think your market share would be in this area?

I think today -- so USB Type-C, we're using an FPGA solution today.

And that's what you're seeing in some of the Microsoft devices that are out there.

It's an FPGA programmable logic solution in there.

The next guys that I'm seeing using that connector are going to be people like LeTV in China and some other people.

So the volumes I don't think are going to be as big as people think as that matures.

But by the time that hits volume, we'll have a different product, which an ASSP that will then move into the FPGA sockets for lower cost for us and higher volumes.

So we're not predicting -- last year, we thought this was a big market.

This year it is probably somewhere around $10 million max, as we walk through this thing.

And a lot of that is just the adoption rate for USB.

Okay.

Good.

Thank you very much.

(Operator Instructions)

Bill Dezellem, Tieton Capital Management.

Thank you.

Relative to your two largest OEMs that you addressed in the press release, would you discuss the weakness that you did see in the quarter and whether the level of softness abated as the quarter went on, or whether it accelerated, and how you are viewing those two customers in 2016?

Yes, if you are talking about the mobile customers being -- one was pretty much right on target with what we saw and the other one was slightly soft, but not really affecting us.

It was like $1 million or $2 million off.

But the bottom line is I think we accurately called those guys in Q4.

I think the disturbing trend is just the fact that in a lot of these consumer platforms, if you take a direction, strategic direction to go after a lower end product line, then you are doomed to compete with a lot of people versus just one big guy.

And I think that's going to be the challenge for some of these people, is profitability more important than leadership?

And I think that will be the challenge for us.

Because again, our products are really leadership products that belong in the highest end smart phones, tablets and other devices within the IOT spectrum.

Thank you.

Sanjay Devgan, Columbia Management.

Hello.

Thanks for taking my question.

Just a quick question on the OpEx trajectory.

I think going into the December quarter, you guys had talked about a $49 million OpEx run rate, and then it came in a little higher than that.

And the assumption was that all of the OpEx or restructuring would realize a full benefit starting in Q1, but it looks like you guys are now talking about $49.6 million in Q1.

And so just the straight line math, you are keeping the full-year guide at $175 million, and I think the assumption was that once you were at an OpEx run rate, we'd be looking at $43 million or $44 million, somewhere in that range.

But given the higher OpEx, the higher base, yet you are keeping the full-year guide unchanged.

Is it fair to assume that the run rate is going to be $1 million or $2 million lower than that previous assumption?

And then as a follow-on, how should I think about OpEx growth thereafter?

We talked about the 175 as being aggregate number on the year.

The number would flux based on mass sets and some variable charges.

So you're starting to see that in Q1.

Q4, we exceeded the target because we had some headcount that remained on books longer than we had anticipated and we had some outside services charges related to the consolidation and systems efforts that were ongoing.

So we expect to still hit that 175.

It will be bouncing around to some degree, based on variability due to mass sets and timing.

Sanjay, the mass sets nowadays can be, on some of our more complex products, over $1 million, and on some of the less complex products, we've been taking a different strategy of using some of the shuttles and other things.

But it still moves around, like Joe said.

And these are $1 million, $1.5 million hits, when you take those.

And related to the leverage going forward, we still see that 175 as a solid number in line with the revenue forecast.

So if we have uptick in revenue, we don't anticipate adding structural OpEx to support it.

There will be some related to sales incentives, but that's really it.

So we think we have a pretty good headroom above that to increase.

There's some variability in there going downward also, but not as much.

Right.

But if I look at -- if I do the new math and if we take the full-year guide, it kind of implies, like I said, $41.8 million on a quarterly basis going forward.

That's better than what I had thought.

And I was just wondering, how should I assume the OpEx ticking down?

Should we start see to see the full run rate starting in Q2, or will there be a gradual decline down and we'll go from an intermediate step and then we'll realize the full OpEx run rate in Q3?

You're going to see a bit more gradual.

Okay.

A bit more gradual.

Okay.

Thank you.

Chris Rowan, FBR and Company.

Hello, guys.

David Haberle on for Chris Rowan.

Real quick, how should we think about royalties from Silicon Image that you collected in this quarter, that you actually collected, and has this business increased or decreased year over year?

Between 2014 and 2015, it decreased.

Between 2015 and 2016, we have modeled it fairly flat.

And between last quarter and this quarter, it will probably come around pretty flat.

Are you there?

Yes, I'm still here.

I apologize.

When we do that, we expect for 2016, the royalty rates actually will probably decline through the year.

And that depends on IP cores and patent sales and things like that.

But today, we've modeled it in, where Q1, it's probably higher and it comes down Q2, Q3 and Q4.

But that's what we've modeled.

We could do better than that, but we're modeling it as fairly flat to 2015.

All right.

And second question, given the Altera-Intel integration, are you seeing any change in the landscape of FPGAs, pricing, increase in sales, anything like that?

Yes, a little here and there.

It's funny, because we jumped into T Cons, that was a direct go head to head with one of those two, I won't tell you who.

And we did pretty well as we walked through that thing.

But I think the biggest changes that we are starting to see is with our distributors just looking straight at displacing on the next design win some of XMA for our product line.

And because they see that some of the customers may not want to deal with Intel anymore or be in a monopoly, if you will, if Intel absorbs Altera and Altera doesn't create the same strategy they did.

So I think we will see it, but it's going to be more of a design win and then out in time, you'll see the benefit.

And there's stuff, like I said, in consumer, it can happen pretty quick.

It can happen in four or five months, if you're doing a programmable (Inaudible).

In some of the communications stuff, it takes quite a while.

We did see some instant pop early on when the announcement came, and we talked to that.

There hasn't been significant change in that going forward.

But we do have a very deliberate plan to replace other companies' pockets, because we believe we have an advantage in the longevity of what we do.

All right.

And last question, you are working down towards $100 million of cash on the balance sheet.

What's your comfortable level there in the cash department?

$100 million.

That's the number that we've stated.

We can go below that.

We've looked at our cash flow and run rates.

We've got coverage for about 7.5 quarters right now, as we look at some of the ratios out there.

There's a mental number that we have in our head around $100 million, so we are definitely looking to optimize and increase as we go forward.

We're focused on cash.

So don't think we're not.

Typically, you aren't.

But again, you want to generate cash and pay down the debt.

And that's a big focus of ours right now, which it should be, right?

When you take on debt, you've got to generate cash and pay it down.

So that's what we're doing.

Right.

All right.

Thanks.

Thanks.

(Operator Instructions)

Francis Chang, 3i Debt Management.

Hello, Darin and Joe.

Thanks for taking the question.

I'm just trying to get a sense of Silicon Imaging's revenue contribution.

If you look at the two years, $366 million of revenue last year and $411 million on a non-GAAP basis, can you give directionally how much was Silicon Imaging during that stub year?

We haven't broken it out that way.

And we've typically avoided talking to it in terms of previous company, current company.

We look at it at different take.

We look at it in market segments, with consumer, industrial and comms.

What we will say is in 2016 we expect Silicon Image revenue to go down because of the transition on the HDTV.

And again, some of that's integration, there's transitions.

And then out in time, we expect it to go back up with the advent of HDMI 2.1, which is the next version of HDMI.

Okay.

Then when you were talking about royalties declined 2014 to 2015, again, could you give me a sense of just quantity of royalty revenue in 2015?

So some of it in 2015, there was quite a bit of stuff in there from IP cores to patent sales -- yes, 2014 -- they had a lot of different things within those ranges and they also had a true-up in there, last quarter of 2014, which caused those things to go way up there.

And that was a couple year true-up that they had.

When we modeled it in, we modeled in for the first couple years to be roughly about $40 million and then we modeled it down over time.

And that could be conservative, but I think over time it should drop from the 40, unless MHL and super MHL and some of the other consortiums we're in begin to pick up.

But we did assume that HDMI over time would decrease and then some of the other areas would increase.

We had it at a $7 million to $8 million a quarter run rate, and it's slowly declining.

Okay.

Thank you.

That's just the HDMI royalty number, MHL royalty number.

Correct.

Thank you.

(Operator Instructions)

David Duley, Steelhead.

Yes, thanks for taking my question.

Just a quick clarification.

You talked about this delayed IP payout or payment.

Could you give us a little bit more details about when you do expect to pick that revenue back up and what caused the delay?

Okay, so I'm going to summarize it real quickly.

We had a contract, they didn't pay us.

We believe they will.

There's always risks.

There's other potential buyers, and we don't have anything material in 2016 listed for that.

That's about all I can say.

Did that make sense?

Okay.

And as far as when you look at all these consumer opportunities that you mentioned, and that's the key driver to this big spike up in revenue in the second half of the year, why exactly are you so confident that those happen and you have such visibility into two quarters out as to why there will be this big recovery?

A lot of it is your interaction with the customer and their interaction with you.

And with a lot of these large customers, they get into the details of what you do for each product that you are talking about and they focus on making sure that you can supply them.

Because again, they don't want to risk their large volume things for some -- for anybody's product.

So you've got to be really careful as you engage in these things that you don't overestimate or under estimate, because both can get you into trouble.

So we've taken the right steps, I believe, to be confident in the assessments we have and the design wins.

There's always risk, as the Safe Harbor says, there's always risk that something happens, but we're really confident in what we have looking forward.

So I guess you have a lot of design wins with large customers, that's the key answer.

In order to move the needle for this year, absolutely.

Okay.

Final thing from me is, you mentioned that you thought the communications business would be down this year.

Could you just elaborate a little bit why you think that is and what the trends are?

Thank you.

Yes.

I think that if you talk to anybody who's dealing with the communications market right now, I think the best I've seen it is flat.

And as we look at it, we saw the LTU build out soften between 2014 and 2015, and then we saw some business analysis by the government, if you will, that seemed to halt some things there in Q2.

And we just don't see anything driving the LTE build-out from China with the macro that we're seeing and the macro around the United States.

I don't see the build-outs.

So we're taking a conservative approach there and saying, we think that it's down.

It's about as simple as we can get it.

And there's always a little bit of a lag with Xilinx and us, because our Q4 communications actually looked okay.

And we're calling Q1 communications down.

They said communications is up.

But again, they typically buy our products first, because they are not expensive, so they'll roll those out and they'll buy Xilinx's, last in, first out technology.

So there's always a difference, and they also serve as a bigger, broader in the wireless than we do.

So if it's a wireless deal, then they may see upsides and we don't.

Okay.

And I want to sneak one more in.

Did you have any 10% customers for calendar 2015, if you could share with us who those were?

We do not.

Excellent.

Thank you.

Thank you.

(Operator Instructions)

We have no further questions at this time.

I would like to turn the call back over to Lattice Semiconductor CEO Mr. Billerbeck for closing remarks.

Okay.

Thank you.

As always, we appreciate your continued support as we drive to grow, improve our financials and pay back the debt.

It is not lost on us that the beginning of 2015 productions fell far short of the stated goals.

We did take quick and decisive action to resize the Company and create long-term shareholder value by making hard decisions fast.

We expect to outgrow the market into 2016, while continuing to execute towards doubling our EPS.

We remain confident in our strategy.

Thanks again for joining us today and we appreciate your support.

This concludes today's conference call.

You may now disconnect.