Seeking Alpha
We cover over 5K calls/quarter
Profile| Send Message|
( followers)  

Cypress Semiconductor (NASDAQ:CY)

Q1 2013 Earnings Call

April 18, 2013 11:30 am ET

Executives

T. J. Rodgers - Co-Founder, Chief Executive Officer, President, Director, Director of Cypress Envirosystems, Director of Agiga Tech and Director of Bloom Energy

Brad W. Buss - Chief Financial Officer, Principal Accounting Officer, Executive Vice President of Finance & Administration and Corporate Secretary

Christopher A. Seams - Executive Vice President of Sales and Marketing

Hassane El-Khoury - Executive Vice President of Programmable Systems Division

Dana C. Nazarian - Executive Vice President of Memory Products Division

Badrinarayanan Kothandaraman - Executive Vice President of Data Communications Division and Executive Director of Cypress India Limited

Analysts

Ian Ing - Lazard Capital Markets LLC, Research Division

Betsy Van Hees - Wedbush Securities Inc., Research Division

Doug Freedman - RBC Capital Markets, LLC, Research Division

Rajvindra S. Gill - Needham & Company, LLC, Research Division

Srini Pajjuri - CLSA Asia-Pacific Markets, Research Division

Sidney Ho - Nomura Securities Co. Ltd., Research Division

John W. Pitzer - Crédit Suisse AG, Research Division

John Vinh - Pacific Crest Securities, Inc., Research Division

Delos Elder

Blayne Curtis - Barclays Capital, Research Division

Steven Eliscu - UBS Investment Bank, Research Division

Christopher B. Danely - JP Morgan Chase & Co, Research Division

Jeffrey A. Schreiner - Feltl and Company, Inc., Research Division

Charles L. Anderson - Dougherty & Company LLC, Research Division

Operator

Good morning, and welcome to Cypress Semiconductor First Quarter 2013 Earnings Release Conference call. [Operator Instructions] Today's conference is being recorded. If you have any objections, you may disconnect at this time. I would now like to turn the call over to Mr. T.J. Rodgers, President and CEO of Cypress Semiconductor. Sir, you may begin.

T. J. Rodgers

Good morning. We're here to report Q1 of 2013, which we believe is the bottom of a long year for us. We'll start out, as usual, with our CFO, Brad Buss, and Chris Seams on the market and back to me for a few questions, and we'll get to Q&A a little bit earlier today. Brad?

Brad W. Buss

Thanks, T.J. Glad to see the markets continue to be more manic than my brother in Canada, so we'll kick in to the good stuff. So again, this is our unaudited results. Take a look at our 10-Q, we expect to file that in early May. We always make a lot of forward-looking languages. They obviously have risks. Please look at our 10-K, 10-Q, press releases and, as usual, we have all the full recons of all our information between GAAP and non-GAAP in the release and on the website.

You'll notice in the press release that we did a little update. We thought we'd refresh our presentation after many years of the same look. We added a bunch of new information, so we hope that you'll find that helpful. And just out of clarity, the Ramtron name no longer exists. It's the F-RAM product line is part of our nonvolatile group, which rolls into MPD. So this will be really the last call that you'll be hearing anything related to Ramtron. But I just want to put a couple things to that, I've read a couple of things where some people were concerned on the Ramtron integration. It actually went very well. It actually became accretive in Q1, which was ahead of schedule. So we're actually very pleased where things have gone there, and you'll see that MPD actually did pretty well in Q1 and that was part of Ramtron. So rest assured, I think that acquisition has gone real well and now we're moving forward.

So for Q1, we had revenue of $172.7 million. It exceeded our guidance of $163 million to $170 million. We were down about 4% sequentially, which was actually quite good because that is actually lower than our normal Q1 seasonality. MPD increased 6%, driven by nDP and Async. DCD increased by 13% in some of the USP area, and they also have our Trackpad group now, which we saw a little bit of pick up. PSD decreased 20% and that was pretty much in line with what we expected. There's a lot of seasonality in the TrueTouch and CapSense business, and we see that rebounding extremely nice in the following quarter. MPD is the largest division that's measured by revenue. Sync is the largest product line, but I expect that to change in the second half as TrueTouch continues to grow every quarter for the balance of the year. By end markets, the handsets, computation and the military decreased as we expected.

We saw good increases in consumer, some portions of communications and then good old industrial and automotive had some nice increases for us, consistent with what you've seen from some other guys. distis were about 74% of our revenue, pretty much consistent with what we've had before. We have our usual one 10% customer and I expect them to grow nicely again into Q2. For the quarter, on a GAAP basis, we had a net loss of $28.2 million, and that was basically some restructuring charges that we took as we talked before about our OpEx base and some final cleanup that we did on the Ramtron stuff. Non-GAAP net income was $4.6 million, that yielded $0.03 per share. That was above the top end of our guidance and really due to the higher revenue and very strong OpEx management, which I'll talk about down the road. If you back out the Emerging Tech Division, the core business actually dropped $0.05 to the bottom line and it was nice to see Emerging Tech at $0.02. If they haven't been at a $0.02 loss for a while; they've generally been $0.03 or $0.04 so they're making decent progress in that arena.

The non-GAAP gross margin was 50.7%. It was down slightly, mostly due to product and customer mix and a little bit of charges for cleanup at Ramtron. If you take out the Emerging Tech, we had a 52.4%, a very respectable number, in my opinion, considering our utilization was down very low, around the 64% range. I expect to see utilization bump up a bit into Q2, and again, unlike other companies that like to keep the factories going and increasing inventory, we don’t believe in that model and so we're going to take some hits in the short term, but we're going to also manage the inventories quite well, which I'll talk about later as well.

ASPs popped up a smidge, and that was mostly due to product and customer mix. OpEx was $80.9 million. It was down 3% sequentially. We're very happy with that because Q1, like most people, we normally pop up due to the increase in payroll taxes resetting and various variable comp plan that reset. We're very committed to managing the OpEx. We've told you before that we expect to keep OpEx on a year-on-year basis flat to down, and that was after absorbing Ramtron. So I think you're going to see a lot of leverage going forward in the next couple of quarters. OIE was $2.3 million. We had a limited interest income, obviously, with the pathetic rates that are out there, and then we had the impact from our revolver out there. Tax was around $342,000, you'll see that working out to about 8%. That's higher than the normal 4 % or 5% we talked about because we basically have some flat tax due to some of our offshore entities.

On the balance sheet, cash was $101.6 million. That dropped about $15.5 million from Q4. We paid out the dividend of $15.8 million. We bought back just under 5 million in stock and we paid about -- the last $15 million related to the Ramtron various expenses, so everything related to the Ram is behind us. Most of our cash is onshore. We had $8.3 million in cash from ops, which is about 5% of revenue, and the free cash flow was just slightly below breakeven, at $1 million. I expect that to be the trough and I expect to see very good cash from ops and free cash flow growth going forward. I do get the odd question on the dividend, payout ratio being high. Obviously, we're at the trough in our earnings so it doesn't look cruddy, but I expect that to increase every quarter going forward. The dividend is sacred here, so you have no concerns, and the payout ratio will be very in line, I think, as we exit the year. Net inventory dollars is $109.2 million. That was down $18.4 million, or 14%. Very, very, very happy to see that. The team worked very hard doing that and not only do we have our inventory down, but our ops guys are executing very well in managing lead times, on-time delivering and quality. So we got happy customers and the CFO is happy with low inventory. We saw just the inventory increase about 9% in dollars, roughly 4% in units and most of that came from our global distis. We saw our independent distis that cover Asia and Japan slightly down for the second quarter in a row. We do expect them to start restocking this quarter as we increase revenue.

Weeks in the channel remained very low, at 6.5 weeks. That's at the lower end of our 6- to 8-week model. Hence, you saw the deferred income pop up a lot in Q1, as did the AR. And that's pretty normal with just the way the fill in the channel went. And I think as you remember, we don't recognize revenue until everything is resold so none of that going on impacted us all in the quarter. Our receivables were up and as was the DSO. And again, that was all due to the timing of the disti shipments. We've collected over $42 million of that AR balance since the end of the quarter. The aging is great and things are good. No changes in the debt or the interest rate. The covenants are all fine at the end of the quarter, so nothing new to report there.

CapEx of $9.3 million, of which $6 million of that came from our friends at Deca Tech. Depreciation was around $10.4 million and actually getting near an all-time low in depreciation. Weighted shares were 145.7 million, fully diluted were 158.3 million. We bought back 411 [ph] shares in the quarter and we ended the quarter with 146.9 million outstanding.

All right. Drumroll for the guidance. So we entered Q2 with a book-to-bill of 1.04, and actually that's been the best all year. I was finally happy to see that sucker turnaround. We're fairly booked for the quarter, and I'd expect to see revenues in the range of $178 million to $186 million. That's up 3% to 8%, and, hopefully, we'll be conservative on that on a long-term basis. Most of the increase will be driven by seasonality and new customer ramps in PSD, with Touch and CapSense being the biggest beneficiary of that. MPD is also expected to increase just slightly.

Gross margin is around 51%, give or take, with the core business running higher. And obviously, that's going to vary depending on customer mix, manufacturing reserves and our foundry mix. OpEx, another good improvement there, I expect it to be down in the range of $79 million to $80 million, and again, like I said, that's going to allow us to drive increasing leverage going forward. Net interest expense was about $1.8 million. Minority interest benefit of approximately $200k associated with our subs, tax expense of about $600k, CapEx of about around $10 million, depreciation just under $11 million, fully diluted stock count, I'm seeing 160 million right now as I expect to see some milestones from RSUs kick in. So you blend that together, we'll get a non-GAAP earnings per share in the range of about $0.06 to $0.08 and that's an increase of 100% to 160% quarter-on-quarter, and obviously a much bigger growth rate than our sales growth projection of 3 8 [ph].

So to sum it up, I'm happy. I'm glad Q1 is behind us. I'm looking forward to a couple of good quarters, and obviously we've got to be cognizant of what's going on in the crazy a** macro but I'm confident that we've hit the bottom and we should start seeing some good things going forward. So I'll turn it over to Chris.

Christopher A. Seams

Thanks, Brad. Let me start with some usual end-of-season and get into the markets and then look at our backlog and booking patterns. In terms of revenue shipments by geography, with the decline in the handset segment, as Brad talked about, Asia-Pacific is still #1 but declined to 58%, followed by North America at 18% and then Europe and Japan are tied at 12% each. We shipped 146 million units in the quarter, and as Brad said, our ASP, mostly from the customer mix and product mix standpoint, actually rose for us to $1.19.

In terms of market segments, Brad talked about the decline in the quarter really being driven by the handset segment. We did get some nice gains in the industrial and automotive segment as the F-RAM revenue rose throughout -- from the last quarter. Looking forward, the handset segment rebounds for us, as Brad talked about, with our latest PSD4 designs ramping, and we also get nice contributions from the computation and industrial segments.

Brad talked about the corporate book-to-bill breaking the unity mark at 1.04. You guys usually ask me, so I'll play by division: MPD was 1.1; DCD, 1.03; and PSD 0.96. They were above unity last quarter. We enter into the June quarter, booked about the same as last quarter, 75% versus 77%. And what's been nice to see is that our backlog has just steadily increased since exiting 2012, and it's continued to do so in the first few weeks of this quarter. And that's even in light of the fact that we've maintained very low lead times in the 4- to 5-week range, so we don't get much visibility. And in that environment, I guess one of the indices I look at is -- Brad talked about the inventory levels of distributors being still fairly low, there's a lot of jockeying for pulling orders in, and we're still seeing a very high level of expedite requests with very low levels of cancellations and push-outs. So I kind of look at it -- I'm looking at a very healthy order environment right now for true in-demand and not inventory stocking.

With that said, let me turn the call back to T.J. for a few more comments on the quarter.

T. J. Rodgers

Some business and technical comments. First of all, I invite you to go on our website and look at the annual report. In that annual report, we introduced PSoC 4. That was not in our plan. It turns out the PSoC 4 project kind of set records for us. What's the PSoC 4 and who cares? The PSoC 1 is the thing we invented in the startup in the early 2000s, and it got us into the microcontroller market. It's not really a microcontroller, but to speak in the common currency, you have to say that. And we penetrated that market and moved from #39 to #7 place in this microcontroller market over the next 5 to 7 years.

We followed on with PSoC 3 and 5, which we followed the market -- we followed our customers' desires by making the analog exceptional. We now have really good analog, best-in-class analog on PSoC 3 and 5. But what we found out, because we're new in this market, is that those chips, which are bigger and expensive, 3x to 5x more expensive than PSoC 1, go into different places. They have a longer design and they have a longer ramp up. And we're happy for being in those new markets, but we decided we needed a refresher on this dynamic, the market dynamic, of PSoC 1. So we took back our PSoC -- we took the PSoC 1 strategy, took the analog from PSoC 3 and 5 and made an exceptional chip with excellent analog, programmable digital, a lot of programmable logic, an ARM core CPU, a lot of memory, all in a chip that we can make for $0.40 and sell for $1. And that's PSoC 4. We think it will be a winner. We also think it will give us a -- not in 2013, but a shot in 2014 at quicker revenue, because this kind of chip tends to go in subsystems, in other words, if you go in the control for a seat, or the air-conditioner, or a rear-view mirror in an automobile, you're in a subsystem and you get fairly quickly to market. If you're doing the engine controllers, the main thing, it's 5 years to get to market. So this product is the most important one we've ever made. It's got more technology on a chip for $1 than anything that's ever been brought to market, anything, anywhere, anytime. And we're very proud of it. So that's all the bragging I'm going to do today.

No, I'm going to brag some more. Toyota. We're in the Toyota. We've been penetrating automotive. And getting in the Japanese automotive, as you might imagine, it's a quality play. We're have -- we have a passion for quality here. We made it into Toyota and we got in a bunch of places in the Toyota Avalon, which is a high-volume vehicle, 7-inch touchscreen navigation system, infotainment controls, all in the center stack, and we also do the audio and climate control systems. So we wish Toyota a lot of luck going forward.

PSoC 4 -- one of the things we messed up in PSoC 3 and 5, is we announced in 2009. I got a keynote address at the Embedded World Conference in Boston and we made a big deal out of it. And we were, to be completely honest, 1.5 years away from having that chip ready. We had silicon, we thought it was working, we thought we would characterize it and take it to market. And the silicon needed an entire revision and even more than one revision. So we embarrassed ourselves. But this time, the PSoC 4 announcement was scheduled to be later, not to do that again, but it worked. This time, we were learning. For example, we had 10 systems, meaning, a motor control system, a temperature-sensing system; 10 systems that our customers would use PSoC in. And when we brought out the first PSoC 4 chip, within 4 weeks, we had 10 different demonstration systems up and running, which is the ultimate test for a chip. And the places where we failed before, where we thought we had a chip but as chip guys, we didn't understand the complications of the end system, and it took a lot of work to get into the systems.

So we've got the systems running. We've got the demonstrations running. We've got the silicon. We're taking orders for the silicon. Another thing we've been criticized for on PSoC 3 and 5 is that our kits cost too much money. Costs a couple hundred bucks to get into the game, and if you want broader distribution, you have to have a cheap kit. You have to think about, like the new standard Arduino, which is sort of a student popular platform. So we've done that. Our new PSoC 4 kit cost $25. So we have a high-end kit for companies that want a design unit at a higher level, but the cheap kit will allow everybody to get into PSoC 4 quickly, which is our goal.

We introduced our Gen5 TrueTouch touchscreen controller. So all of the companies that compete in this market go through generations and then if you miss a generation, like we did last year, you get clobbered. Our touchscreen revenue went down $100 million year-on-year because touch -- excuse me, TSG4, Touchscreen Generation 4, wasn't in the market on time, and we lost market share to our competitors. The recent TSG4, which I bragged about a year ago, didn't make it. It wasn't the chip. The chip today is the way it was a year ago. We're proud of it today as we were a year ago. We are starting over 1,000 wafers a week, and for those of you who can do calculations, you can go figure out the revenue on that. So TSG4 is becoming a success for us a year late. Problem was we didn't understand the system well enough and we didn't understand the firmware for a 32-bit ARM controller well enough. We do now. We've reorganized. We have a new software center that's a world class software center. I have an Executive Vice President who is a software guy only, who sits on the executive staff. That's new for us. So we've become, let's say, professionals at the software, at this level. This ain't Microsoft, this is not Java, but the software is still complicated enough that you have to have a competent organization to run it. So we turned on the software and TSG4 is now starting to meet its promise. The good news is that being chip guys, we think about every generation as being a new chip, and we started TSG5 immediately. And TSG5 is smaller, more economical and it's better for noise than TSG4. The main -- one of the main -- there are 3 things that now differentiate products in the touch market. Two are features, waterproofing. Can you be in fog or rain and will your touchscreen work? The second thing is glove. Can you be in the northern climate in the winter and not have to take off your glove and the cell phone works? Those -- we have both of those and we believe we have best-in-class solutions.

The other thing, which is not a feature but it is a problem we have to solve, is that -- I've talked about this before, when you charge your cell phone, you plug into a 110 outlet, you go to the USB plug and you charge through your USB plug. Many people like to work with their phone while they're charging or they may have their phone on their desk and charge it while it's under their desk, so when they go in their car or go out, the phone is fully charged. Many phones, many times don't work. The touchscreen does not function in that environment or it's flaky. If you -- I think you've probably seen that yourself if you have a smartphone. The reason for that is chargers that are used create noise that noise, which ranges from 0 to 40 volts, gives a signal that perturbs the tiny little signal coming from your finger and screws up the way the touchscreen works. We have solved that problem. We have the best-in-class solution and I'm absolutely sure of that. And that's Touch Screen Generation 5 and we've got that chip up and running. It's another success we had in our chip group last year, which had a great year. While we were suffering in financial hell last year, our chip group had a great year and TSG5 is ready. So we've got a rapid time-to-market follow-on chip to TSG4.

On the touchscreen side, we won Huawei. One of their high-end phones, it's called Ascend Mate, and winning -- the winning features were thick gloves and water rejection. ZTE, the other big Chinese cell phone company and the fourth largest cell phone company in the world has also selected TSG4, which I told you has ramped before, for a 5-inch, 1080-pixel phone called Nubia Z5. On the technology front, one of our startups, AgigA Tech, makes very high-speed, high-density, battery-powered memories. We've been in the non-volatile RAM business for a long time. We started this company, which makes memories in the gigabit scale, as opposed to the megabit scale or the kilobit scale which our other memories are. These are system-level memories not chip-level memories. They're currently one of our startups. They have had revenue. They're starting to ramp revenue. The purpose of this is to point out that they are a finalist in the Edison Awards, and the Edison Awards are the Oscars of our business and we're proud of that.

And finally, we already talked about it, we paid a cash dividend, that's effective today. When Brad was talking about this cash dividend, I did a calculation. As Cypress's second-largest shareholder my dividend, I receive yearly, is 6x bigger than my base pay. So when he said the dividend was sacred, he was not kidding.

Brad W. Buss

You were already underpaid.

T. J. Rodgers

I'll be glad to be what is called a managing shareholder.

Okay. We're ready to take questions.

Question-and-Answer Session

Operator

[Operator Instructions] Ian Ing, you may ask your question and please state your company name.

Ian Ing - Lazard Capital Markets LLC, Research Division

It's Lazard Capital Markets. So gross margins, you mentioned, Emerging Tech is a bit of a headwind. But also PSD, I mean, is this sort of a new level of gross margins here or is this somewhat transient? And what's driving that?

Brad W. Buss

I think it's a combination. I mean, remember when the fab is very underloaded with memories being mostly externally done now. The burden of that is falling heavily on PSD and then secondary would be DSP -- DCD, I should say. But yes, I mean, the Touch margins have come in, as we've talked about over the last year. So we really kind of need the growth to come back, which it is starting to come and we're seeing very, very nice growth in the Touch, consistent with what we expected in Q2. And I think as the fab loads back up, we will get there. But they're not going to be running, nor have they ever ran, in the high 50s, so I wouldn't be expecting that on a long-term basis.

Ian Ing - Lazard Capital Markets LLC, Research Division

Great. And then looking at your bookings commentary, it looks like you're guiding turns orders in the mid-20s now. Recently, you've done closer to the low-20s and you got high-20s last quarter. So perhaps talk a bit about are you getting more confidence on turns orders and longer lead times? Or what's going on there?

Christopher A. Seams

This is Chris, Ian. Yes, I did talk about it. We've had some quarters where we've done higher than the mid-20s and while we're hopeful for that, again, with the short lead times and the way we're getting these bookings, I can't really see the last half of the quarter yet. So we're hopeful, but our guidance is our guidance at this point. Yes, I definitely feel better.

Ian Ing - Lazard Capital Markets LLC, Research Division

Okay. And last question, MPD revenue, up 6.2%. You said portions of the comms market are up, I mean, perhaps you could drill down. Is that networking or telecom? If it's telecom, is it wireless or wireline?

T. J. Rodgers

Networking, in general, is up just a little bit on both the wireless and the wireline side, but a lot of that growth was driven by our newly acquired F-RAM product line. It's ramping up quite well. They've got a very diverse customer base, and now that the customer base is seeing that Cypress is fully putting the full force of their development and product portfolio out there, they're coming back and starting to buy those products. So that's ramping up for us. That's what mostly drove the growth. Ramtron.

Operator

Betsy Van Hees, you may go ahead and please state your company name.

Betsy Van Hees - Wedbush Securities Inc., Research Division

Wedbush Securities. I wanted to go back to the question on the Memory Products Division. So last time, you guys talked about inventory -- that there's inventory correction going on with the Ramtron business, and a lot of growth came from this -- in Q1. So as we look at Q2, are we getting back to the run rate that Ramtron used to be at? Or where will we get to that? They used to be around $14 million a couple of quarters ago before you acquired them. So I was wondering if you could talk about where Ramtron is and then how the Memory Products Division, SRAM, will be tracking in Q2 compared to the guidance?

T. J. Rodgers

Okay. So on the Ramtron side, yes, there was an inventory correction and that's -- that inventory is starting to deplete in the field, and for that reason, the revenue went up. I think it still got room to grow for the rest of the year. We don't quote the actual revenue by the product lines but let me just say that directionally we'll be back to pre-acquisition revenue rates within the next quarter or 2, and it's growing quite nicely.

Brad W. Buss

Yes. And, Betsy, remember, keep in mind, they were on a sell-in basis, and they had a lot of distributors and they had a lot of inventory.

T. J. Rodgers

And I think your second question was related to SRAMs. Can you repeat the question?

Betsy Van Hees - Wedbush Securities Inc., Research Division

Well, just going back to that comment on structure. We've seen, unfortunately, a lot of blowups in that group, so -- and there's been a lot of concern in terms of what type of growth we're going to see from infrastructure spending in the back half of the year. So I was just wondering, the leading supplier to that sector, how you guys are seeing things in Q2 and how -- what you're hearing from your customers? And then I just have one more question on gross margin.

T. J. Rodgers

Okay. So that is going to be flattish to slightly up. Of course Europe can't get much worse than it was last year, and that's starting to come back a little bit. But the big wildcard is China infrastructure spending. And I have a -- a lot of that's dictated by the decisions that their government makes, which I can't even predict what our government is going to do, other than raising taxes continuously. So predicting what the Chinese government is going to do is even harder for me to do. That'll be the key factor in the second half of the year.

Betsy Van Hees - Wedbush Securities Inc., Research Division

Okay. And then congratulations on the OpEx control. Gross margin guidance, 51%. When are we going to kind of get back to the run rate that we were, in the mid-50s? Are we going to start to see that in the September and December quarter? Or how should we be modeling as we exit the year?

T. J. Rodgers

I mean, I think you'll see it go up every quarter throughout the rest of the year. I mean, like I said, the utilization, we need to get the bump from that. And then the comm is a big portion, I mean, we have decent gross margin in that area and as it continues to be fairly tepid, as that business picks up, that's going to come and add to it. But I don't see us being in the mid-50s or high-50s this year at all, right? We need that revenue level to get back up, we need the fab full. And then we need a decent portion coming through the foundries. When we hit those mid-to-high 50s gross margins, we are doing 40%, 50% through the foundry, our fabs were full and the MPD was blowing and going. But I think pushing up towards that mid-50s is going to be, kind of, the next leg, and then we'll go from there.

Brad W. Buss

I have 2 comments, one on Ramtron and one on the SRAM business. I'll do them in the reverse order. I have analyzed, with the best data you will find, because we're 40% share and therefore we really know what's going on in the market. I've analyzed the last 1.5 decades of SRAM business in the annual report and there's -- what I modestly consider to be a very good analysis of SRAM dynamics, and the fact is SRAMs are going down. They've dropped -- or was it compound? I think it's 15% per year over since 2000. The reason for that is that ASIC can integrate more and more SRAM and even more importantly, the design methodologies used today allow companies to integrate SRAMs with less and less risk. So the business has dropped down to the $400 million TAM. In 2000, just for your information, the TAM for SRAMs was $6 billion. And even after the crash, the TAM for SRAMs was like $3 billion and it's now, like, $400 million -- excuse me, it's $600 million. We're at $400 million. So there's a long-term trend in SRAMs. Now that's perturbed by things that make you forget the long-term trends. For example, when personal computers took off in 1995, they all had 4 SRAMs in them, went through the roof. Then there was a crash in '96. When SRAMs went in and routers, in the 2000 time frame, SRAMs went through the roof. There was another jump and then of course, routers started doing integration and they haven't completed that yet. So there's still a lot of router business. And then there was another bump for cell phones. So every time some new system takes off and they've got time-to-market pressure, they don't want to take the risk of making an SRAM, especially if it's an exotic, difficult-to-make SRAM. For example, the SRAM fused in routers today, that radiation tolerance is equivalent to what the space SRAMs used to have a decade ago. Because there are tiny amounts of radiation that make it to the earth, and you literally can see a difference in failure rate in SRAM in Colorado versus New York City, literally. And as a matter of fact, the reason we discovered that radiation leaks [indiscernible] to the world is that IBM, a decade -- 2 decades ago found that their computers had a higher failure rate in Denver than they did in New York City.

And because they did the pioneering work, the neutron radiation rate is normalized to the neutrons per square centimeter per second in New York City. There's a little factoid you probably didn't know you're going to learn today. But the point is, how you design those SRAMs, how you design the cell, how you architect them so that a single radiation hit creates enough errors to be corrected and not too many errors overall to be corrected, all that stuff is the big deal. How do you make an SRAM? Does it run on a watch battery for 3 years that'll hold data is a big deal? So there are a bunch of SRAMs, where although generic SRAMs can get [indiscernible] and ASyncs, so a bunch of SRAMs that only real SRAM people can make. And we believe the SRAM market has pretty much declined to the residual, where specialization can allow us to maintain market share. And we're going to grow market -- we can maintain a market and it will -- our goal is to grow market share within that market. So that's the SRAM story. We're hoping that the decline we've been living through year-by-year for 1.5 decades is about over, and therefore the growth we see in PSoC, which largely has offset the SRAM. So as I've said, we've been flat in the revenue, losing on one side and gaining on the other. Our PSoC growth can show up on the top line. That's the SRAM story. The Ramtron story, we don't comment on business units. We don’t -- each business unit, of course, has competitors, and it's extremely important for competitors to know a given business unit, so we just -- we won't talk about a given business unit, so a company competing with one of our divisions has our key information. I'll do a onetime mini data dump on Ramtron for you because I think it's reasonable to do that, since you spent $135 million buying it for us. They were, as you said, $14 million per quarter rough-cut, so that's $56 million a year. They weren't making money. They were not well-managed, and they were bloated. And they didn't deserve to be a company. They are actually 2 clicks down in our hierarchy. They're a business unit, okay? And that's the right thing for them to be. Therefore, we got rid of all the corporate overhead. We literally cut the expenses -- the operating expenses at Ramtron in half in 1 quarter, from $30 million to $15 million a year. And now they're making money. And if you look on an incremental basis, so that is if you look at incremental revenue from Ramtron and incremental employees and costs in buildings and rent and taxes we inherited, we're making 20%-plus profit on that incremental transaction. Of course, we loaded back Cypress people into it, to integrate it, to bring in functions that we're understaffed there. They get corporate taxes, so, yes, they got rid of being a company, but they have to pay taxes to our company and pay for their share of our G&A expense, et cetera. Net result is last quarter, they made money. So we never had a loss quarter for Ramtron is -- with a partial quarter, but the first quarter of 2013 was a fully loaded profitable quarter and we were accretive. And on an incremental basis, they were highly accretive. And yes, we're -- we will hit back towards that revenue rate, and we will achieve that, I believe, this year. We're not there yet, but I believe we will. Next point, we expected in Telegraph, a very big revenue drop. We came in and found out the way they were selling was basically dumping products to distribution, jamming the channel. And they were taking revenue credit for shipment to distribution, which is a known bad practice in our business. So we had to reverse all that, start taking credit only for revenue shipped from distribution to end customer. We -- they had 20, 30 distributors. 30 is a nightmare of distributors. We had to terminate a bunch of distributors. All those distributors had the right to put inventory back on us. We took back a bunch of inventory. We have a bunch of inventory now, 200-plus days? We've over 200 days, but we can burn that off in 200 days, okay? So that's not that long to wait to burn off the inventory. The good news was we overestimated the "inventory correction," and I think that's a misnomer. Because they were selling the distribution, because they had jammed the channels, there's a huge amount of revenues of inventory and distribution. So people buying Ramtron RAMs bought only what they needed, only when they needed it, and they were quite sure they could get it, and if they couldn't get it from distributor number 40, they'd get it from distributor number 20. Therefore, their customers are hand to mouth. So we took all those inventory back, and all of a sudden, bam! they need SRAMs next week. So it turns out, we were over 50% higher in Ramtron revenue relative to our estimations that we're going to have a quote inventory correction. We didn't really. We basically had an inventory correction in the sense we took back a lot of inventory, but basically that flow-through just kept on going till it was signed for. It's a fine acquisition. It's going to make money. The more I look at the technology, the better I like it for ultra-low power and by the way, if you read the annual report, there's a mini-section in there I wrote on comparison of nonvolatile SRAMs and what makes for high speed, what makes for high density and what makes for high power. Ramtron has got the lowest power of any nonvolatile memory. [indiscernible] says nonvolatile RAMS are the fastest of any you can buy. And then our AgigA Tech that I talked about earlier RAMS are the biggest nonvolatile RAMS you can buy. So we've decided that, that's a high-profit niche market. It's -- we think it's a 25-plus percent pretax market. We've done that with our internal efforts, and Ramtron is going to be an important part of that. So it was a nice tuck-in acquisition. It's already integrated. We like the guys; we like their technology. And that's the Ramtron story. And then we're -- like I said, for competitive reasons, we just don't want to talk about a given business, because there's 4 or 5 companies out there that will get information they really shouldn't have.

Operator

Doug Freedman, you may go ahead.

Doug Freedman - RBC Capital Markets, LLC, Research Division

RBC Capital Markets. Can you talk about what your expectations are for TrueTouch for the balance of the year in terms of revenues? In the past you've talked about a percentage growth target for the year, are you still comfortable with that same range?

Hassane El-Khoury

Yes. Doug, this is Hassane. I run PSD. We're still on track when you look at the expected growth for the remainder of the year, between 15% and 25%. Obviously, that's many macro issues, but we see a strong funnel in the design that we discussed earlier in the highlights section also will help drive that as well.

Doug Freedman - RBC Capital Markets, LLC, Research Division

Can you offer us some color on where that is going to come from, is that dependent upon China handsets, are you targeting some U.S. or European Tier 1 companies for touch solutions? Any color you can offer would be helpful.

Hassane El-Khoury

I'd say primarily it is the China-based or APAC-based design win. We talked about the DTE win that we've had that's China-based. That's -- a lot of the 15% to 25% growth will happen from the APAC region.

T. J. Rodgers

Then we do have a couple of new customers, as we've talked about before, in Europe, and we're picking up some other share with some of our other larger current incumbents, so. It's quite pretty nice. I mean, the thing I like is, it's not reliant on just 1 guy. There's 3 or 4 opportunities that we have, so if there is a delay or one of the phones doesn't take off, you're not betting the whole farm, but nothing's in the bag until it's done. And the nicer thing, unlike last year, you're going to see some of the growth in Q2 and Q3, not all in the back half. And hence, the reason I think you're going to see him still grow pretty nicely quarter on quarter.

Doug Freedman - RBC Capital Markets, LLC, Research Division

And lastly, can you provide us an update on any of the legal issues regarding GSIT or any other pending dates that we should be aware of?

Brad W. Buss

The update is the same as the last time. The ITC case is being reviewed by the full commission of the ITC, and we're expecting a decision at the end of June.

T. J. Rodgers

The incremental change is we sued them in the ITC. We believe and currently do believe that they violate a number of our patents. That number is growing every day as new patents get issued. The ITC voted that they did not infringe our patent and -- or the ITC judge ruled that they didn't infringe our patents. We believe that is wrong. We -- then GSI did a maneuver to ask the ITC to look at the validity of the patent, so their game was, "Let's get the patents invalid." The ITC ruled that all of our patents were valid. So if you stand back and look what the ITC did, there was a very quick trial and they voted not to intervene at all. That means this matter will get dumped back into the federal court systems. We sued them a couple of years ago in Minnesota, and that's where we will pursue it. And I am -- of course, in any legal issue, you always get surprised, you can get surprised. But I am highly confident that I can show in the courtroom, and this time, it won't be our lawyers and experts in front of an ITC administrative law judge, it will be me and our team in front of a jury in Minnesota, and I'm absolutely confident that I can convince the jury and the judge that they are violating our patents and they don’t have any patents of their own. They're not big on IP protection, as I said earlier, I personally have more patents than their whole company does, and therefore there's nothing to fire back at us, so we've got this pokey antitrust case going on. I'm not worried about that. That's where we are.

Operator

Raji Gill, you may go ahead.

Rajvindra S. Gill - Needham & Company, LLC, Research Division

Raji Gill, Needham & Company. A question on the top line growth for 2013. Kind of, what's your expectation now for the annual guidance? You talked about that you're going to -- earnings will grow more than the top line. Any kind of indications of what the top line for the year is going to be also a low compare in '12?

T. J. Rodgers

Raji, we don't guide the year, as you well know and, trust me, every quarter has enough volatility. And it's like my weight, it's up, it's down, it's up, it's down. So I would say, we definitely dropped. I think we gave pretty good guidance, but looking at the screen, for some reason, someone's not liking something. But I think when you look at where the industry is going with the economic backdrop, I think we'll start growing out good or better in these next few quarters versus the average market, though. So it's not going to be a 10% to 15% a quarter, but I think it's going to be very solid compared to what I think you're going to see out of most guys. And more importantly the leverage coming back, right? I mean, we feel that on the way down, but as you saw the last time when we came back, I mean we were definitely going to be delivering substantially more earnings than the top line. So I think we really need to take it quarter by quarter in this macro world.

Rajvindra S. Gill - Needham & Company, LLC, Research Division

It seems like -- in 2012, you did $0.55 non-GAAP earnings. Your revenue has got to grow a little bit more in order just to be flat, given the fact that margins are coming back as strong, so I'm just kind of getting a sense of -- yes, you talked about [indiscernible] faster than revenue. I don't really see how you're able to get there unless OpEx comes down further.

T. J. Rodgers

I think you'll see some level of OpEx, but I mean, not a ton. I mean, I think all the incremental revenue and gross profit dollars that come with that, you'll pretty much be seeing the majority of that go to the bottom line.

T. J. Rodgers

You'll see the OpEx come down, in investor terms, on the order of $0.02 to $0.03 per quarter.

Rajvindra S. Gill - Needham & Company, LLC, Research Division

Okay, got it. The TrueTouch business, I'm glad that you kept the 15%, 20% revenue growth. Can you talk a little about the ASPs and the mix shift that's happening? You're targeting China, which is the bulk of your design wins. But the ASPs there are probably a lot lower, and so the units must be pretty significant. Just can you maybe talk a little bit financing units ASPs in China?

Christopher A. Seams

Hey, Roger, it's Chris. I wouldn't characterize it as China. I would characterize it as big accounts, and there are some big accounts now in China. And so with those big accounts comes an expectation on pricing. So I'd say China, frankly, is not much different than the rest of our big accounts, either throughout Asia or based out of Europe. It's a one world as far as pricing goes now. Clearly, T.J. talked about what we've come out and finally gotten to market with TSG4 and what's now out there in TSG5. And clearly on those parts, when the customer selects them, the 2 notices that are out today are very high in handsets. I think the [indiscernible] is a 5-point something and the Huawei is a 6.1-inch phone. Brad laughs because it hangs out of my pocket when I carry it. But those command a higher ASP, so at the same time we are prospering in the -- what used to be the feature phone market. Now it's the mainstream smartphone market, the 4- to 5-inch range in Asia. We do have high-end phones as well, and the mix has been -- I'd say it's been the same price reduction that we've had over the last 2 years that we see now.

Operator

Srini Pajjuri, you may go ahead.

Srini Pajjuri - CLSA Asia-Pacific Markets, Research Division

CLSA. Brad, just the gross margin guidance. The last time we saw 180 handle on the top line was, I think, back in March of 2012, and the gross margins were much higher. I'm just wondering to what extent this is mix-related versus utilization versus any other factors.

Brad W. Buss

It's really a combination of both. I mean, again, part of it the memory group, down from where it used to be, and that tends to be the higher margin stuff. And the utilization is down as well. Because remember, the memory guys, they moved out of that fab every 6 months for the last 1.5 years, they've been going fab more business to UMC.

T. J. Rodgers

[indiscernible] utilization internally, obviously, we've got a fab weighting you down and it's not filled and ours hasn't been for the last year, that's a problem. There's even a smaller effect externally. When you deal with 4 or 5 foundries like we do have a significant group. You've got to take care of yield. You've got to have a VP level guy. You've got a production control. It's not equivalent obviously to a fab, but you got a group like that; it's required. And then as you ramp on your foundries, and your foundries ramp way down, you've got some fairly high overhead on a small amount of volume going through your foundries. So utilization is the number one culprit. Growth will fall-through it. Like we said, we cut ourselves down so we made money in the $172 million quarter, and we need to run the top line up. We have a little bit more cutting to do; it's already planned. It's staged. And we know when it's going to happen. It can't happen immediately. So we've got a little bit more. We go in a couple of pennies a quarter, cutting costs, but after that, it's top line. And the top line right now depends. Ramtron is part of the top line story and recovery from our mistake in 2012 in TrueTouch is the other half of the story, and that's what's going to drive our revenue this year.

Brad W. Buss

And again, remember, I mean, we took our inventory down 14%, and I expect it to be flat to down again next quarter, even with revenue growing nicely, okay? So you're seeing a lot of other guys, the other analog and microcontroller guys, tend to just keep building all the way through and they pork out on inventory to keep that utilization high and keep stuff on the balance sheet. We don’t believe in doing that, so you've got to keep that in mind as you look forward and you compare with what other guys are doing. I think what we're doing makes a lot more sense, especially in this very low lead time environment.

Srini Pajjuri - CLSA Asia-Pacific Markets, Research Division

Makes sense. And then, could you remind us what the utilization number was for the quarter?

Brad W. Buss

Mid-60s. I think it was 64, 65, I said.

Srini Pajjuri - CLSA Asia-Pacific Markets, Research Division

Okay. And then, Brad, finally, you said you're expecting couple of good quarters. Could you please remind us what your normal seasonality is for Q3?

Brad W. Buss

Q3, 6% to 9%.

Operator

Sidney Ho, you may go ahead.

Sidney Ho - Nomura Securities Co. Ltd., Research Division

Nomura Securities. So last time, you guys have a book-to-bill at above 1 was Q1 last year, and your backlog coverage at that time was, like, 85%. And now you're -- it's above 1 and backlog coverage is a little lower. How would you compare the current environment than less than a year ago?

Christopher A. Seams

Sidney, it's Chris. I would say we're probably -- that's of our own doing. T.J. and our operations group has completely restructured the way that we service and supply. And at the same time that we've done all that, they have reduced our lead times down. Frankly, most of my products I sell are in the 4-week range, and somebody calls up, I can do something special for him to get inside of that. And we've kind of trained our customers that's the horizon for them. So if you just take 2 weeks out of our run rate, you'll get kind of that change in backlog and coverage versus where we were a year ago.

T. J. Rodgers

And again, the manufacturing guys are probably bringing down cycle times a few more days in the second half of the year, which is very good.

Christopher A. Seams

And while I suffer from the lack of visibility by a few weeks, trust me, I love having the fact that I can react faster than the other guys in taking order off the street.

T. J. Rodgers

Nothing is driving the inventory down. It's a planned thing. I've spent a lot of time on it. We have a very confident operations and manufacturing group right now that's doing better than we've ever done. One is, I want to take the business off the street, so I want to be able to quote 4 weeks all the time. And I want to be able to do better than 4 weeks for somebody who needs more product right now. That would just get you more share in your account. Two is, we were writing off on the order of $0.01 per quarter forever on inventories that has aged and we finally gave up on it because we're never going to ship it. The mechanism for that in the standard terms and conditions in our business, our orders can be given any time, but they become noncancelable and non-returnable in 30 days. So if you can make a product and get it to market in 30 days, you can only pull [indiscernible] and ship them for orders that are NCNR orders. If your manufacturing time is longer than that, then you run the risk of less than a cancellation. The way our customers work is they kind of order more than they need, and then right before the 30-day mark, they trim the order to what they actually need. Sometimes they go up, sometimes they go down, but there are ones that go down. And then, we get a dump of inventory and usually, we can sell it, but every now and then, we can't because it's a special part or a weird part or it's got a different mark on it, whatever, and we can't sell it. So our goal is to get our manufacturing cycle time inside the NCNR 30-day window. When you do that, your total inventory required -- just think about it as a pipe. You got a pipe that's delivering $200 million a quarter, and if your manufacturing time is 4 weeks, you've got enough inventory to sustain 4 weeks' worth of that business. And if you go from 6 to 4 weeks in your pipeline, your inventory goes down. So we're driving inventory down methodically. I have board level goals that I get graded on and right about every quarter to do that. So it's good for customers and it's good for us, and it eliminates the source of waste. It's been about $0.01 a quarter and most likely, we don't get that $0.01 back.

Sidney Ho - Nomura Securities Co. Ltd., Research Division

Great, that's very helpful. My follow-up question is, can you talk about the linearity for the quarter and how things have progressed so far in April? I think one of the analog companies said this week that orders decelerated in the month of March, and that seems to be kind of the opposite of what happened last quarter.

Christopher A. Seams

Sidney, Chris again. Well, their orders must have come to me because mine have not decelerated. Like I said, it's been a methodical trick out of the low December. It's just been nice and steady up into the right in terms of backlog coverage and growth and linearity, in terms of orders and shipments, it's very, very high in the high-80s to 90s right now, early in the quarter.

Sidney Ho - Nomura Securities Co. Ltd., Research Division

Got it, got it. If I can squeeze in one more on CapSense. Can you remind us how big this bucket is within PSD, what kind of typical ASP for this in, say, a smartphone, and what kind of margin profile do you -- does it have?

Christopher A. Seams

No comment.

T. J. Rodgers

I'll give a couple of facts on CapSense. Last time I gave you facts, I watched the little red line go downward, but just between the truth and watching the red line go down, I got to say the truth. CapSense are copper buttons and sliders and pads. So the buttons -- an electronic button is the obvious thing. On the far end of CapSense, you have -- it's like a trackpad where you've gotten x, y array of copper. So CapSense is like TrueTouch. The difference being CapSense is copper on PC boards and TrueTouch is a clear electrode material on glass so you can see through it and see the screen in your cell phone. We've seen this play before. We have 40% market share in capacitive buttons. The average selling price for these buttons is under $0.50, fact one. Fact two, the gross margin in that group is near 50%. Fact three, the profitability of this group is above 20%. And one of the things I need to warn you about a little bit is I'm not knocking gross margin, and if I do regression analysis on corporate valuation, price to sales ratio, for example, versus gross margin, there's a high correlation. Higher gross margin leads investors to believe that there is high value-added products, there's more fall-through to the bottom line, et cetera, and all of that is good enough. So at some businesses we have, where the gross margin is lower, but the operating expenses are also very much lower, and the net result is you can run, pick a number with a 48% gross margin and still make, pick a number, 24% pretax profit because once these markets mature, it's basically manufacturing, 5 guys run the business and ship it, and you're shipping in the tens of millions of units. This, I believe, is the transition that TrueTouch is going through right now. And 5 years from now, I expect to sit here and tell you about the TrueTouch business and how we won in that 40% market share because we can make stuff and we can ship tens of millions and we have high quality and good technology, and that's the dynamic that has already happened to CapSense, and we want it. We're making great men in CapSense, but that's what's going to happen in TrueTouch. When I told the truth of where it's going last time, everybody was, "Oh my God, commodity market, push the button." Your call. But that's what's happening. We know how to play that game, that's because our DNA has RAM guys, and RAM guys live that all of their life. And we will prevail at the end of this journey on TrueTouch.

Operator

John Pitzer, you may ask your question.

John W. Pitzer - Crédit Suisse AG, Research Division

It's Credit Suisse, guys. Brad, I apologize if you've touched on this, but over the last 2 quarters of the balance sheet, just a lot of variability around receivables. Is that all a function of distribution mix both down in December and then up big here in March, or can you just give me a little bit...

Brad W. Buss

Yes, exactly. I mean, what -- deferred income and the AUR all tend to go in the same direction. And it's just the timing because we had a lot of shipments that the distis were taking, getting positioned for the growth in Q3. They went out mid-to-late in the quarter, so the receivables go up because they don't pay us for 30 days. And like I said, I mean, north, I think, like $45 million of that has already been paid off in the quarter as the terms came due. So no concerns there. I mean, our AUR has been just spanking good for years.

John W. Pitzer - Crédit Suisse AG, Research Division

And then, Brad, you guys talked about a full year target on TrueTouch, and I think you also mentioned it will be growing faster than the overall business sequentially in the June quarter. Did you quantify how much faster TrueTouch would grow in the June time frame?

Brad W. Buss

Much faster. Much higher than the top end of our guidance, put it that way.

John W. Pitzer - Crédit Suisse AG, Research Division

And then, guys, as the smartphone market, which is a key platform for touch, becomes more dominated by kind of the Chinese OEMs, can you help me understand the expected margin profile you'll get on a Chinese OEM mid-tier device versus, let's say, a U.S. OEM high-end device, or how do I think about that dynamic?

Christopher A. Seams

John, Chris here. Yes, I kind of answered it on an ASP basis. So I really see it more as a volume play versus a geography play. I mean, I -- there are low-volume guys in, I'll say, China that are willing to pay big dollars because we'll support them and give them features that they can't get just on their own or buying from [indiscernible]. So it really is more volume. If we pay attention and have the resources to pay attention to a smaller volume account, obviously, we're going to get higher ASPs and margins. But characterizing geographies in the world, I don't really see it. I see volumes of the world. And that mid-tier market probably is the most brutal, and you've heard our comments about where we think the margins can be long-term from T.J. just now.

John W. Pitzer - Crédit Suisse AG, Research Division

And then, guys, lastly when I look at the other bucket inside of PSD, that's the non-TrueTouch bucket. Clearly, not down as much from peak levels vis-a-vis TrueTouch, but still down significantly and kind of a resumption back to where you were maybe 4 quarters ago, which still represent really strong growth. I'm just kind of curious, is the growth in that bucket all a reflection of our view of the cyclical upturn for the industry, or what other specific drivers or how do you guys see the recovery in that other part of PSD?

Hassane El-Khoury

This is Hassane. We see it not just in TrueTouch, obviously. It's also related to the CapSense growth that we have seen already. There is another element contributing for our platform PSoC, which is the non-CapSense and non-TrueTouch, which is part of PSD, that's more on a longer but steady growth given the end markets, the nonconsumer end markets that we play in, driven by both industrial and, as T.J. mentioned earlier, with a highlight on automotive. So we're looking at a long-term steady growth on our baseline, with industrial coming in first and then followed up by automotive, in that order.

John W. Pitzer - Crédit Suisse AG, Research Division

Is there a multiple to MCU growth, industry growth that we should think about for that other bucket?

Hassane El-Khoury

I think the growth for platform PSoC that we're seeing, if I look at it in relative to the industry growth, we're looking at definitely 2 or 3x the growth of the industry.

Operator

John Vinh, you may ask your question.

John Vinh - Pacific Crest Securities, Inc., Research Division

It's Pacific Crest. Just a clarification on the 15% to 25% growth for the year for TrueTouch, is that inclusive or exclusive of CapSense?

Brad W. Buss

Exclusive.

John Vinh - Pacific Crest Securities, Inc., Research Division

Great. And then my follow-up is, you guys obviously seem very confident about TrueTouch Gen5. Where is your confidence level this year on potentially getting a high-end smartphone design win with Gen5 on a kind of non-China kind of high-end smartphone this year? Where do you guys stand on that?

Christopher A. Seams

John, it's Chris. I'll say that we've got 2 offerings in Gen5. One is aimed squarely at the, I'll say, the mass market, which is the 4- to 5-inch range, so I don't think you'd consider that a high-end non-Asia because those are typically now above 5 inches. I get laughed at when I talk on my phone now because T.J. actually said it looks the old Motorola brick when the handsets first came out. But I think in terms of the high-end segment, we've got a version of Gen5 that will be out in the second half of the year and will certainly be in design and you probably won't see it to the market until the first half of next year.

T. J. Rodgers

Mobile World Congress in February of next year, how many phones will be selling that will add Gen5 system?

Christopher A. Seams

My answer is, more than it take -- I'll have to take off my socks and shoes to count. There will be more than I can count on my hands.

T. J. Rodgers

We're in designs and designed into a bunch of phones right now. Gen5 is a special chip. And like I said, that charger noise that I explained in detail, which also by the way is explained in the annual report how it works, is becoming a big deal if you look at the blogs, where users now with a net can start complaining about features on products they find out, that even if the company goes to the trouble of shipping a good charger that -- you can do that. You can spend more money in a charger and has a noise recharger. Even if you do that, you buy your one charger, or get your first go into your phone, then you buy your second one to put in the second place. And you buy the RadioShack one or you buy the highest volume chargers. And if you look at them all of the high-volume chargers in the world, they have noise in the 15- to 40-volt range, and they all screw up cell phones. And we have solved that problem. I think, as I said, we've done our best job at it. And that's the level right now of getting design wins, and we are getting those wins.

John Vinh - Pacific Crest Securities, Inc., Research Division

All right. And then my last question is in a quarter, what was approximately the mix of smartphones and tablets in TrueTouch? And then how do we think about that mix for the full year?

Christopher A. Seams

John, it's Chris. For the quarter, it was 90-plus percent handsets, and that will change a little bit in the second half of the year as the eReader business picks up again.

Operator

Glen Yeung, you may go ahead.

Delos Elder

This is Delos on behalf of Glen with Citi Research. I just wanted to ask about the opportunity for your 5th gen touch controllers. In particular, can you revisit how the risk of integration of a touch controller with the LCD driver factor into that? And do you address it?

Hassane El-Khoury

Yes, sure. So at the end of the day, it is the customer choice of what stack to use. But I can tell you, our generation, our Gen5 specifically is very well-suited. We have already engagement with customers, and with the LCD partners to enable our touch controller to drive their in-cell or on-cell depending on which way they would like to go. But we're very well-suited and are actively in that market today.

T. J. Rodgers

That is right. And we can parse the thing 2 ways. Today, almost all cell phones have a thin film censor that is on top of the LCD display, and that's called on-stack. And then the trend is going in-stack, meaning the sensor for touch is buried in the -- inside the stack. And then there are variations of that called on-cell and in-cell beyond the scope of this conversation. So on-stack is where everything goes today. In-stack is what people are talking about. We have an instance, we have had in-stack designs. They are much more difficult for our customers because the noise generated, the LCD-noise generated in-stack is pretty substantial. That the idea is we're going to get lower cost of ultimately because the -- when the sensor goes away it becomes part of the LCD, the incremental cost of the LCD will be lower than what you pay for a sensor. Everybody that I'm aware of today is driving a sensor. The Samsung LCD still drives it with a real touchscreen chip. They do not have a single chip that drives both the LCD and the sensor. That integration is postulated. I don't see it. The reason I don't see it is the technologies that are optimum for driving LCDs are very different from the technologies that are optimum for sensing a sensor. The sensors want analog. They want a microcontroller, et cetera. And the sensor in technology that drives the LCD is typically Dynamic gram-based technology. The analogy would be how come Dynamic RAM is not Intel chips? Why doesn't Intel just make a bigger chip yet and put the Dynamic RAM memory on the chip? And the answer is that Dynamic RAM technology is very different from the technology Intel needs to make a microprocessor, and it would actually cost more to put a Dynamic RAM on an Intel chip than buy one separately. In effect, yes, you take all the masses of microprocessor and all the masses of DRAM, which are very different, and add them together, and the extra mass make both the microcontroller and microprocessor and the DRAM more expensive. I believe that Dynamic applies to the LCD driver and the TouchSense chip. And furthermore, as TouchSense chips, our lowest-end TouchSense chip today, they're low capability for the cheaper phones, but it's $0.50. So where is the benefit of taking all the risk of integrating those 2 chips with disparate technologies having 2 engineering groups doing totally different jobs and trying to get all the stuff on one chip? Where is the payoff for the chip against the $0.50 chip? So we will obviously keep an eye on it, but I just don't see the integration of the chips happening. I do see the integration of the sensor into the LCD stack happening. That actually makes the sensor chip's job tougher, and puts more stress on the sensor chip, means you need a higher order, more sophisticated sensor chip, and not one that's lower order that you can stick into the other chips.

Unknown Analyst

Great. I just wanted to ask one more question about MPD. For the past several years, it's been a pretty stable quarterly run rate, yet profitability has shot up probably from the 20s to now, looks like, 50 to 60 range. So I'm just wondering, as far as how much more profit you could squeeze out of this market? Has the low-hanging fruit been tapped? Or what do you see as kind of the near-term driver?

Christopher A. Seams

Glen, it's Chris. Just to clarify, you said M, as in Mary, PD?

John Vinh - Pacific Crest Securities, Inc., Research Division

Yes, sorry, the Memory division.

Christopher A. Seams

Okay. Well, we're always running -- we're running at pretty high margins, but we have an engine that continues to drive cost reduction. I mean, it's in our D&A. We're bringing out a new family of Asynchronous RAMs. These are going to be the best Asynchronous RAMs in the world, and they're going to have a better cost structure and better quality and better performance. So if you go -- you're right. If you go back 7 or 8 years, we've got a gross margin trend up, and it just comes from a continuous improvement culture that we've got. But it has to tilt [ph] at some point. I mean, I don't think you can expect step function increases in the future.

Operator

Blayne Curtis, you may ask your question.

Blayne Curtis - Barclays Capital, Research Division

Barclays. I just wanted to better understand the TrueTouch guidance. Given what you guided for Q2, it seems like the back half would be almost a double or a 80%, 90%. It seems very similar to last year. And typically, seasonally, Q4 is not that strong. You have a large customer who typically balances inventory. So maybe if you could just walk through kind of your thoughts? And you should have a pretty good view as the Q3 ramps as well? And is this year different seasonally?

Christopher A. Seams

Blayne, it's Chris. Yes, obviously, the second half of the year is going to be bigger than the first, as you noted. There's 2 dynamics. One is the normal seasonality of programs that are trying to hit a certain kind market releases. The other is that, frankly, we're just going to gain share back as we get TSG4 designs finally out to the market. That's what's happening. And as TSG5 ramps, we're just going to take share back. So that's the second dynamic that you add on top of typical seasonality.

Brad W. Buss

But then again, Blayne, some of the newer customers we've talked about, they begin their ramp in Q2. And obviously, the second half is going to be bigger than the first because when you go a negative in the first quarter, I mean mathematically it's tougher, but we have a lot more ramping in Q2 than we normally do. So that spreads it out a little more.

Blayne Curtis - Barclays Capital, Research Division

Great. And then a question for Dana. On the -- with Ramtron being, I think you said the majority of the growth in the March quarter, I guess the core SRAM business didn't grow that much. But with Ramtron coming back faster, you would think you'd see a little bit better growth into Q2. That's kind of point one. And then two, as T.J. had kind of outlined, you're kind of at the bottom of this secular headwind. Why isn't the core SRAM business picking up a little bit better?

Dana C. Nazarian

Well, the core business is going to be up in Q2, so I'm not sure where you got that impression. We'll be up.

Christopher A. Seams

The challenge really more, Blayne, is we need the comm business to come back, right? I mean, everybody is waiting on the great, gold comm coming back, which you know -- who knows? You can only run the network so hot, and all this merger stuff flying around hopefully settles down. And I think when people start spending, as you all well know, we'll do well in that area. So we're cautiously optimistic on it. I mean, we get more and more data points from the suppliers, which we deal with all of them, but we want to see the orders, too.

Blayne Curtis - Barclays Capital, Research Division

And I just want to clarify, maybe I misheard that Ramtron came back faster in the March quarter. Does that trajectory not keep going, and you get closer to the prior run rate in Q2?

Dana C. Nazarian

It keeps going. I think the percentage increase in Q2 will be from Q1 to Q2 will be smaller than from Q4 to Q1. It will continue.

Operator

Steven Eliscu.

Steven Eliscu - UBS Investment Bank, Research Division

UBS. First question. In the first quarter, I would like to better understand what drove the USB growth, considering computation is typically seasonally down in the first quarter. And as a result, how should we think about DCD revenue tracking through the rest of the year?

Badrinarayanan Kothandaraman

My name is Badri and I run DCD. So the growth in DCD was from 2 factors: one is the USB, which I'll talk about; and the other is Trackpads. So finally, USB 3.0 is getting to be a decent fraction of our overall revenue. We have shipped over 1,500 development kits, have over 600 customer engagements. We expect to grow sequentially through the year and more so in 2014. And the product we have is called the FX3. It basically connects any general-purpose interface USB 3.0, and it's highly programmable. This serves a very broad range of applications in arranging permission mission, 3G gestures, documents, scan and light-tracking systems, surveillance and the digital [indiscernible]. So that was part of the reason we grew. And coming to Trackpads, like what we said in the last confidence call, we expect Trackpads to double in 2013 versus 2012. And the first quarter was quite strong, with Trackpads exceeding our plans by almost 40%. Those are the 2 reasons.

Steven Eliscu - UBS Investment Bank, Research Division

All right, that's helpful. Switching to the annual; report, T.J., that was extremely helpful. The specific chart I want to focus on is with regards to the SRAM market. And just with the 15% CAGR decline embedded in that was a relatively stable period where we bounced around $1 billion for it looked like around 6 years. And then over the past 2 years, we've dropped by on the order of 1/3. So in addition to integration, do you think that some of that shift has to do with maybe move to -- a move to other memory types, such as RLDRAM? And as you think about what you should do going forward, given integration is something you're just going to have to live with, should you be pursuing some of those high-speed DRAM technologies or perhaps partnering with Moses or something like that, that puts you at a leading edge of a very large memory that's a lot less susceptible to integration?

T. J. Rodgers

Okay. That's really a good question. First of all, I kind of, due to the forum, gave the 50,000 [indiscernible] as the decline of the SRAM market. The bank component is integration, but also you pointed out that shifting to other kinds of memories has been part of it. To give a quick example, router, we used to have -- router is a complicated thing, and there are like 5 or 6 different SRAM sockets in a router. And SRAMs used to populate all of them. There are some of them, like, they're buffers for the data, where, because you put the data in sequentially and take it out sequentially, you don't have static random access memory. You don't have to bounce around to get something at random. You can count on the fact the data's going to go in, in the queue and come out in the queue. And those sockets transferred over to DRAM. RLDRAM, so yes, we -- anytime another memory can replace an SRAM. The SRAM is the highest performance best powert speed product, speed product memory in the world, period. The only thing is it's not nonvolatile. And that's, of course, our nvSRAMs are that next step. And -- but they're expensive, and it's 10x the price for an SRAM bit -- even more than 10x -- the price of an SRAM bit for a DRAM bit. And that's based on square nanometers of silicon required to make a bit. So anytime a customer can figure out a way to replace an SRAM with another RAM, he does this. The SRAM he's got left, it needs a sophisticated customer. It's a Cisco, for example. Then they can figure out, in some cases, assuming the SRAM is not an odd one or a difficult to make one hwe will change it over. The reason there was that the current flat period, which gave everybody, including me, the illusion that the world was going to flatten out of $1 billion and stay there. So I believe that one of the downward trends was offset by one of the spikes, where there's a spike of business that picked up and then went back with integration. I believe we are near the end of the trend. As I look at the RAMs we make, and I see that we're probably 2 years away from where the RAMs we make aren't RAMs that you'd want to integrate. And we have direct conversations with people, where they say, "Well, with an integrator, we can buy your RAM." And we say, "Well, let's explain about how RAMs work in this environment. Let us give you our textbook, which is 200 pages thick, on the interaction of various kinds of radiation with SRAM memory bit, and explain to you about all the little problems you might think of running into." And that invariably convinces them to say, "Okay, we get it. We don't want to get in that business. You guys are smart guys. You've been doing it for 30 years, and we get it, that that's not something we can just use some ASIC RAMs, slap them in and then hope it's going to work. I think we're almost there. I think there is a residual minimum. In your -- the last question is, do I want to get in DRAMs? Hell no, I don't want to get in DRAMs. I spent my career working on Moore's Law. Moore's Law is one that's exciting, but you don't make money out of Moore's Law. And as the managing shareholder of the corporation, I have money to invest in R&D, and DRAMs would be the last place. You look at -- I admire Micron. They're a great company. I've always liked them, but I don't want to be in that market. I have many, many more interesting places to be in. And now that we've got PSoC technology-- just think about PSoC and the fact that it is a microcontroller surrounded by programmable everything, okay? And think about that as a product, and I can go to some company that makes programmable radios or a radio. I can go to a company that makes chargers. I can go to a company that makes these little companies and I can say, "Gentlemen, you give me an IP block. You design it. You drop it into a PSoC. And all of a sudden, you've got $500 million infrastructure that has a complete product that's firmware-programmable, that has got software -- designed software for your customers. We can take you as opposed to you try to take an idea in a block and turn it into a company. We can turn that into a product that's better than one you could ever make in your entire career of your company. We'd like to start a business with you and just kind of think about of a billboard. And then the whole billboard's surrounded with PSoC blocks. And in the middle, it says your IP here." And that to me is a much more exciting possibility for all of the new things we can get into quickly, cheaply than hammering on memory. I do not want to buy another software that costs more than a Boeing 737. I've been there, I've done that. It's not fun. I agree 100%.

Steven Eliscu - UBS Investment Bank, Research Division

Well, that's a great segue to my last question here on PSoC 4.If you can help us understand the decision process of going from where you are on PSoC 5 with your ARM-based products pushing down to the Cortex-M0 versus perhaps pushing to the higher end with the Cortex-M4, where that segment of the market perhaps would be more willing to pay a premium for your analog capability.

T. J. Rodgers

The question states things in an either/or way, and we don't view it that way. We created a franchise at the low end, with 8 bits. And a franchise, like I said, settles from number 39 to number 7 in 8-bit microcontrollers. We penetrated the market that has united something special we never ever, 25 years later, would have penetrated. The PSoC 4 is to keep that franchise alive, and to put something into that franchise that nobody else can touch. So basically, we have the discussion about the base SRAM business going down, the PSoC business growing, with some of the 2 being flat, okay. I don't want that to happen again. So I'm pumping some money with a few chips into the low-end so that I can maintain that 8-bit base. Now the "8-bit" really means cheap. And now we can make a 32-bit microcontroller literally for less than a penny. So cheap now means the 32-bit microcontroller. So within the next year, you'll have a button, and you'll put your finger on the button, and that button will sell for less than $0.50. And what you don't know is underneath that button will be a 32-bit microcontroller that will make the signal and the noise ratio and the parameters of that system better than any button that ever gotten made literally in the history of the world. So it's a -- I am shoring up the base. I don't want to build my skyscraper in quicksand so every time I add a story on top, I sink a story, which has been the problem with SRAMs. So where PSoC 4 is going to bolster the bottom and absolutely in the future, all PSoC 7, I've been working on it for 3 years, and we haven't talked about it at all yet, but absolutely, we will look at the other end as well.

Operator

Christopher Danely.

Christopher B. Danely - JP Morgan Chase & Co, Research Division

This is JPMorgan. Just to close up on the SRAM. Would you care to give us your sort of yearly prediction for growth there, revenue growth?

Brad W. Buss

No. Chris, I mean we're not doing it in total on other than the touch, which has been a very touchy area. We haven't thrown out any numbers. I mean, we'll expect it to grow, but, again, a lot is going to depend on the comm. I mean, we're drinking the Kool-Aid like everybody else, expecting the comm to come back and to actually let him grow nicely in the second half, hopefully, and have a good exit rate for the next year.

Christopher B. Danely - JP Morgan Chase & Co, Research Division

Real quick on OpEx. So great job in Q1 and Q2. How should we expect OpEx to trend in the second half, conceptually?

T. J. Rodgers

Flat to downwards.

Christopher B. Danely - JP Morgan Chase & Co, Research Division

Okay, great. And then last question, is there anything we can sort of model in for your incremental gross margins as revenue goes up? Or is it in a bit dependent on the mix with foundry plus your own side?

Brad W. Buss

It's the customer mix. I mean, again, I see it trending up a point or 2 kind of within that range, but it's a little hard to model out every quarter right now.

Operator

Jeff Schreiner.

Jeffrey A. Schreiner - Feltl and Company, Inc., Research Division

Feltl and Company. I just want to stay with the kind of philosophical, I guess, questions here. And one question I want to kind of ask T.J. about is, I think that Steve was starting to push towards this a little bit, but how are you going to need to adapt your current SRAM portfolio? And the way I say adapt is, does there a need to be a shift in the type of I/O 1you're using and maybe move much more to serial I/O base networking memories as the speed in the network continue to increase over the next few years? What's your thoughts on that?

T. J. Rodgers

We are in the next generation of memory. The I/O is changing. The whole product is changing. So we're -- all of the end customers are looking to not run into the problem that T.J. was talking about before when he was using the analogy of the microprocessor being saddled with all the extra costs to get the memory onboard. And they're moving towards 2.5D solutions and 3D solutions. These are solutions where you put 2 chips side-by-side and connect them with a piece of silicon instead of a printed circuit board. And when you configure it that way, you start to change the I/O and interfaces between the 2. So that's the first step. Then the step after that could possibly be a high-speed serial interface more engaged with both big FPGA vendors, the ASIC guys and the ASP guides, as we develop the roadmap together. My short answer for the question would be both. We're going to change. You have to change the outlook because you have to get more gigabytes per second through the memory. And there are 2 ways to do it. You take a given number of output, and you're running faster. And we're talking factor 10. So you go from 1 gigabit per second per output to 10 gigabits per second per output or even more. Or alternatively, you take your output, say it's 32 bits wide, and you make them wider. And therefore, you're running with the same speed, but you run more of them, therefore you get more data. We are seeing interest in both. We used to talk about memories as being 1-bit wide, 4-bits wide, 8-bits wide, now 32 bits wide. Some of the router memories are 72 bits wide. I can tell you for a fact we are actively working on memories that are 1,152 bits wide. So you can get a whole lot of data out of 1,152 I/Os. There's so many IOs, you can't get in the package anymore. And that was the trend for the so-called 2.5D integration, 2.5 dimension, where big chips will be mounted facedown on bumps on what are called interclosers. You can think of them -- not think of them, they're exactly silicon printed circuit boards. And the reason for doing that is you can get finer lines, and the expansion and contraction of the printed circuit board and silicon matches the chip, and you could put bigger chips on. That's the trend. That trend is standing towards 1,152 bits wide. There's even an industry standard developing. It's called HBM, High Bandwidth Memory. On the other side, if you look at the FPGA guys, they've gone the other way. They have 10 and even 40 gigabits per second pins that zoom data out at super high speed. Of course, the problem here is you get to funnel a lot of data into one pin. You've got to run a pin at super high speed and make the data go on and off the chip without making mistakes. And then on the other end, you've got some firehose [ph] coming at you. So you've got to fan it back out so you can get it to the usual speed again. And both are happening, and I predict we will have both in the future. We're working on both as we speak.

Jeffrey A. Schreiner - Feltl and Company, Inc., Research Division

Okay. And for my second question, Brad, I was just hoping to dig in a little bit on the PSD. I think the question was asked earlier, when you're looking at the year-over-year variance in terms of the gross margin for PSD, and you had talked about touch coming down. But for a lot of competitors touch is still, I think, certainly on the mobile side, where some of your strength is, is at this point, it's still over 50%. So given the year-over-year decrease you've seen in the gross margin for PSD, could you quantify at all how much is possibly utilization versus maybe margin compression?

Brad W. Buss

Well, like I said, I mean, the group that takes the biggest branch of the SAM under utilization is PSD. So I mean that's definitely a substantial portion of it. But obviously, with the ASPs that have came down in the past and until the newer products kick in, the gross margins are down on a year-on-year basis. But I would say utilization's number one on the cradle.

Operator

And our last question comes from Charlie Anderson.

Charles L. Anderson - Dougherty & Company LLC, Research Division

Dougherty & Company. So I wanted to focus in on TrueTouch. You gave, you rated the guidance for 15% to 25% revenue growth. I wondered if we look at gross profit dollar growth and operating income growth, if it would be equivalent or if there'd be any degradation there? And then thinking about next year, with the new chip, how you think those trends will manifest themselves, revenue growth versus profit growth?

Brad W. Buss

Yes. But, Charlie, we're not putting them out obviously for competitive reasons as well, but they'll be moving in the right direction.

T. J. Rodgers

Our gross margin and profit in TrueTouch like that as a company will grow at a higher rate than our revenue growth. We -- this quarter really is the bottom when we're reporting to you pretty much for all of the profit indices. We've swung down the base of the company from the impaction [ph] overhead and the SG&A overhead. And we're going to get dropped through on the gross margin line even on the leanest of our product lines.

Charles L. Anderson - Dougherty & Company LLC, Research Division

And are you, just to kind of drill down there, would we have a dynamic where we're having decreasing margins with decrease in ASPs and you're getting the lift from decrease in OpEx associated with TrueTouch? Would that be the dynamic that's playing out?

T. J. Rodgers

We will have increasing revenue, which will make increasing gross margin and increasing pretax profit dollars in TrueTouch. And the fall-through, the rate of growth of the fall-through numbers, gross margin and gross profit -- and pretax profit will exceed growth of revenue. That is, we will not offset our fall-through by ASP degradations.

Charles L. Anderson - Dougherty & Company LLC, Research Division

Got it. As you think about the design wins that you're getting now for TSG5, does it look like you're going to have a better ASP profile into '14 versus what you're seeing in '13?

Badrinarayanan Kothandaraman

I think the ASP degradation is behind us. If you heard what T.J. was talking about, the Gen5 specifically. That generation's superior than any other competitors. The customers know it. The customers are willing to pay for it because at the end of the day, it does save on their system level. If you can pass the noise performance right from the start, it gives them a time-to-market advantage, as all the superior perception from there and the customers as well. So we are looking at it, like Chris said before, it's really a volume thing, but we are maintaining healthy ASPs in general because of the value and the future said that we, hand like the glove and the waterproofing that T.J. covered earlier.

T. J. Rodgers

Our pitch is, yes, we know the price is going down. Yes, we know there's a terrible price pressure on the cell phone. If you like the cell phone to work with a crappy charger that most people attach to it, we can do that for you. Our competitors cannot, and we want a dime for that. And we think it's worth a dime. And that's the dynamic. First of all, nothing lasts for more than 2 quarters in the business. But for the next couple of quarters, that's where we're going to be.

Brad W. Buss

All right. Thank you, everybody. Look forward to seeing you on the road.

Operator

And thank you. This concludes today's conference call. You may go ahead and disconnect at this time. Thank you for participating, and have a good day.

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.

If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!

Source: Cypress Semiconductor Management Discusses Q1 2013 Results - Earnings Call Transcript
This Transcript
All Transcripts