奇景光電 (HIMX) 2006 Q2 法說會逐字稿

Good evening, ladies and gentlemen, and welcome to the Himax Technologies, Inc. conference call. (OPERATOR INSTRUCTIONS). As a reminder, this conference is being recorded.

It is now my pleasure to introduce your host, Mr. David Pasquale of The Ruth Group. Thank you, Mr. Pasquale. You may begin.

Welcome everyone to Himax's second-quarter 2006 earnings call. Joining us today from the Company are Mr. Jordan Wu, President and Chief Executive Officer, and Mr. Max Chan, Chief Financial Officer.

After the Company's prepared comments, we will have time for any questions. If you have not yet received a copy of today's results release, please call The Ruth Group at 646-536-7003 or you can get a copy of the release off of Himax's Investor Relations website.

Before we begin the formal remarks, the Company's attorneys advise that certain statements in this conference call including statements regarding expected future financial results and the industry growth are forward-looking statements that involve a number of risks and uncertainties that could cause actual results or events to differ materially from those described in this conference call. Factors that could cause actual results to differ include general business and economic conditions and the state of the semiconductor industry, level of competition, demand for end use application products, reliance on a small group of principal customers, continued success in technological innovations, ability to develop and protect their intellectual property, pricing pressures including declines in average selling prices, changes in customer order patterns, shortages in supply of key components, changes in environmental laws and regulations, exchange rate fluctuations, regulatory approvals for further investments in our subsidiaries and other risks described from time to time in the Company's SEC filings including its Form F-1 dated March 13, 2006 as amended. The Company undertakes no obligation to publicly update or revise any forward-looking statements, whether as a result of new information, future events or otherwise.

At this time, I would now like to turn the call over to Mr. Jordan Wu. Please go ahead, sir.

Thank you, David, and thank you, everyone, for joining us on today's call. I will now review Himax's performance during the second quarter of '06 and discuss the outlook for the third quarter. Max, our CFO, will then provide further details on our financial performance.

Now, let me start with the second quarter. We are pleased with the results we were able to achieve in the [54th] quarter for the TFT-LCD industry. Revenues and EPS came in within the range we have guided.

Despite higher than normal inventory levels and reductions in production activity by certain of our customers during the quarter, our net revenues grew by 54% year over year and declined slightly by 2% sequentially. As we expanded our business with several [large panel adventurers] in South Korea, Taiwan and China, our revenues from our related parties increased 29% sequentially. Revenues from related parties declined 21% after a strong first quarter. We remain the principal supplier of display drivers to CMO. Overall, we think our more diversified customer mix puts us in a stronger position now than before.

Revenue from small and medium-sized display drivers grew 213% year over year and declined 2% sequentially in Q2 as demand from certain model display customers slowed down and the market was characterized by significant ASP erosion. In spite of the challenging market environment, we continue to make good progress with several world-leading small and medium-sized panel makers. We have designed into several module suppliers of [PON] handset customers and customer electronics panel makers. We're confident that this design effort will materialize in the near future.

Reflecting industry conditions, positive pressure was higher than normal during the quarter. The large panel driver prices experienced up to a high single-digit decline in certain products in the second quarter. (indiscernible) was packaging, in particular, suffer from higher degree of pricing pressure, reflecting a more significant decline in the price of gate material.

Small and medium-sized displays drove up prices sale by around 10% sequentially. In spite of manufacturing cost reductions, we were able to achieve the [foreign SP] did have an impact on our gross margins, which declined by about 2.3% sequentially.

Our operating margin before share-based compensation declined sequentially to 11.6%, comparable to the level we saw in the second half of '04 when the panel industry was working through inventories and experienced a significant price decline. GAAP diluted EPS was $0.10, which includes share-based compensation of about $1.1 million or $0.01 per diluted share. Based on -- or generally 8.5 million diluted weighted average shares compared to $0.07 in the second quarter of '05 and $0.12 in the first quarter of '06.

Now let me turn to the outlook for the third quarter. We are seeing the overall TFT-LCD business environment improving from the second quarter. We expect the business environment to improve later in the third quarter as customers benefit from the typical seasonal off time and work through the remaining inventories. PC and TV-related demand should improve sequentially. Our small and medium-sized display drivers revenues may decline in the third quarter as certain of our customers continue to experience inventory correction. Specifically, we expect to have revenues in the third quarter of '06 to grow at the mid-to-high single-digit rate compared to the second quarter of '06. We expect our gross margins will remain flat or decline slightly.

Our non-cash diluted EPS before share-based compensation is expected to be at a similar level as start of the second quarter. However, we expect diluted GAAP EPS to be in the range of $0.03 to $0.04. This includes a charge of share-based compensation of around $12 million to $14 million or $0.06 to $0.07 per diluted share. The significant increase in share-based compensation is due to the new grants of restrictive share units, which is expected to occur at the end of September '06. Next, our CFO will be providing more information on the RSU later on.

Now let me turn it over to Max Chan, our CFO, for further financial details.

Net revenues in the second quarter were $171.7 million, representing year-on-year growth of 54% and a sequential decline of 2%. Large family freight driver revenues declined by 1% sequentially. Small and medium-sized display driver revenue declined by 2%.

While our top line in the second quarter hasn't changed much sequentially, the revenue breakdown by application did show some notable changes. PC-related display drivers, including display drivers for monitors and notebooks, accounted for approximately two-thirds of our total revenues in the second quarter and increased from around 60% of our total revenue in the first quarter.

On the other hand, TV driver revenue declined to approximately 20% of our total revenue in the second quarter from about 25% in the first quarter. Revenues from small and medium-sized display drivers remained at the low teens percentage of our total revenue.

We saw revenue growth in all geographies except Taiwan, which was impacted by the decline in shipment to Chi Mei. The gross margin had declined more than we had expected to 19.2%. This was primarily due to higher than expected ASP erosion. We continued to target unit cost reductions to offset ASP declines. Also, we're making good progress in qualifying new foundries to further diversify our supply base. We continue to closely manage expense controls during the quarter with operating expense excluding share-based compensation increased 1% sequentially to $13 million. At the same time, we continue to invest and expand our human resources, increasing our total headcount to 811 as of June 30th.

Share-based compensation was $1.1 million or $0.01 per diluted share in the second quarter. We intend to make our annual grant offering [city share] units to employees at the end of September. We expect the fair value of the 2000 RSU -- 2006 RSU grant to be in the range of $23 million to $26 million, of which approximately $11 million to $13 million we will vest immediately on the paid-up grant. The remainder will vest equally in three installments over three years. As such, our share-based compensation in the third quarter is expected to be in the range of $12 million to $14 million or $0.06 to $0.07 per diluted shares.

We have recently decided to amend our vesting schedule in such a way that employees who are granted relatively small amount of shares will have a large portion of their granted shares vested immediately. We believe the change in vesting schedule will make our RSU a more competing program to attract and retain talent.

For non-operating items during the second quarter, we wrote off our $1.5 million investment in [Lymasters] as the Company has filed a bankruptcy. Interest income increased significantly to $1.8 million in the second quarter as a result of $148 million in net proceeds raised in our IPO. Net [fairy] in change gain was $1.4 million in the second quarter. In the second quarter, Himax generated a net operating cash flow of $15 million. Our ending cash balance stands at $167 million with no debt. Capital expenditures for the second quarter were $4.1 million.

Operator, that concludes our prepared remarks. We can now take any questions.

(OPERATOR INSTRUCTIONS). Heidi Poon, Piper Jaffray.

This is Heidi Poon calling for Tore Svanberg. Could you provide a little bit more color on your Q3 guidance? What kind of linearity should we expect for Q3 in terms of by segment the shipment pattern?

I don't think we're seeing major changes of our product mix compared to second quarter. I will probably, firstly, comment by giving more details on the customer [breadth sale]. I think we said in our comments earlier that related parties, meaning CFO, the share is starting to increase. In Q2, the CMO share decreased to around just short of our total revenue. This compared to over 60% throughout the entire of last year. It was quite a major decrease for us.

Finally, we also commented that on related parties, meaning primarily other large panel customers, they came in strongly during Q2. Acquired for Q3, we are seeing CMO picking up again the volume acquired for certain of our customers. I can't probably provide the specific details on any specific customer's business. But certain of our customers -- I will just give one example or two examples where we are the sole source in certain major projects of theirs. And because, as you are aware, the [TSE] industry is suffering from a difficult market environment and particularly in certain product segments. So that has slowed down substantially or in some cases, this production in certain of their product areas where we have not been the sole source provider.

So, although CMO came in strongly in third quarter, we do have a few other customers who also came in strongly in the third quarter. We do suffer from major declines from a couple of our other large not CMO customers. So, lump sum together, that sort of explains why we are guiding for an increase -- however, not a substantial increase from Q2.

In terms of product mix, I think we feel we are still enlarging our market share for large panel; although, certainly that has been a very difficult -- we have seen a very difficult business environment in that sector. But I think -- in the small/medium-sized panel, we're becoming the first source or the predominant source of -- for a couple of [worthy then] providers of small panel [push], who have seen before in our [East] history.

But instead, we saw in particular a few others, where we're working very hard to design to their projects. It does take time for us to see major volume. In other words, we're trying to ship most of the world's for the first-tier panel makers, serving first-tier handset makers or consume electronics makers. That is the process we just need to go through. So we are not saying we can see the top-line or bottom-line results immediately, but we feel quite good about the progress in terms of design-in and penetration to such customers right now.

So are we talking about maybe Q4 timeframe that we might be able to see some of the volume from these design wins?

I think Q4, we're starting to provide more detailed comments and guidance next quarter. But I will just say we are confident we will continue to see increase from Q3. How much increase, I don't really think we have the visibility now to provide detailed comments.

You mentioned earlier that your operating model now, it's akin to second half of '04 when the industry is going through inventory adjustment. But what is your target operating model going forward, especially as the expectation for a business to improve from here?

I think in the expense side, we will continue to invest in R&D. You see, we're still increasing the headcount. I think that is crucial to our future business development. Our revenues should increase through the second half of the year. While we continue to diversifying our foundry base, designing new projects and hiring new people.

So overall, we're seeing better momentum in the second half of the year. But on the operating side, we're hoping to see some economy of scales and cost reductions now at our core site. While in the revenue side, the ASP pressures continues but definitely not in the same magnitude as we have been experienced in the second quarter of this year.

Frank Wang, Morgan Stanley.

Can you talk about what's the shipment growth you're expecting for large-sized and small/medium-sized for the third quarter?

We're seeing kind of a PC-related, namely monitor in notebook demand picking up. So these shipment and revenue will grow sequentially for the large panels no doubt.

While for the small and medium-sized, including the handset in consumer electronics, and we are -- there's certain inventory adjustment and control of [see already] demands in certain of our customers. So the overall -- the shipments in the large panel display drivers will increase significantly. While in the small and medium-sized, especially in the handset side, that may slow down -- still slow down. We are expecting a picking-up in the fourth quarter. But we think the second quarter and third quarters, that's the inventory adjustment in certain of our handset customers.

Given the fact that the bulk of our business today is still large panel drivers. So your question, I guess, more or less reflects our view of the shipment for the whole Company sequentially -- shipment growth. We've guided earlier for a mid-to-high single-digit top-line growth. I think I implied probably 3 percentage points higher number for shipment growth, i.e., I think you will be booking above a low teens percentage point kind of growth shipment-wise.

I made some commentary earlier regarding why you don't expect a substantial third-quarter revenue growth. Will you please repeat that comment again?

Our guidance, again, for the top line on sequential basis, is a growth of 8 to high single-digit points.

Can you maybe [frame] the panel customers are probably looking for a double-digits revenue growth in the third quarter? Can you talk about the delta between their double-digit revenue growth versus yours and what's the likelihood that maybe things will be back-end loaded and maybe actually things can improve a little bit from here on?

We certainly hope so. Firstly, I think our experience in the past two quarters tells us that in this kind of industry environment, sometimes we just can't see the degree of visibility we would like to see. So we do get an urgent pull-in or pull-down of orders from customers -- [DS] or large-panel -- large panels. So I guess you are right in the sense that we are presently hoping for certain upside. However, the guidance is the guidance. So I'm not trying to suggest that you guys should be targeting for higher upside. Certainly, we don't exclude that possibility.

Having said that, though, I said earlier certain of our customers have fear going through their inventory correction. Many of them are trying to actually -- in the past, they have suffered from higher than normal industry inventory levels. Now they are trying to achieve a low and "normal" inventory level. That is certainly a negative impact for component suppliers.

I think also, I explained earlier certain customers, certain major models of theirs. For profitability reasons, they have decided to build out substantially the shipment, meaning changing their capacity in relation to other product areas. The area which has been filled out in one or two cases happens to be our purpose to have Himax being the sole source supplier. So that certainly is also a major hit for our top line.

But, again, I think I will be lying to you if I say we do have very good visibility at this point in time. Because to tell the truth, the industry has been changing so much, so quickly that the (indiscernible).

Regarding the RSU grant of $23 million to $26 million commented earlier for 2006, how do you derive that number?

We try to set it in (indiscernible), which actually we said to the masses during our IPO that we are trying to maintain a long-term distribution of around -- I say around, meaning it's not precise -- but around 30% of our profit of that same calendar year. Meaning on September 30th, we are going to make the grant for this year. At such point in time, we are going to make an estimate of our whole year '06 profit before share-based compensation. We will take around 30% and make a distribution as bonus to our employees.

Again, we are -- certainly at this point in time, we are not certain about the whole year's profitability. So it is out of our judgment of the whole year profitability and not get down by 17% and we have 30%, which is the number for our estimated RSU grant of today. Then we take that, and we used to say the grant will then be expended in four equal installments because of the vesting schedule. In the third quarter of each year, you will see a pick-up of a quarter of that year's grant. But we have made certain amendments not very long, also as a request from our employees where certain employees, they may be too new to the Company or they may be very junior, so they are not getting a lot of shares to begin with. So you divide that shares up to four installments. It becomes very, very small for them in each installment.

So I think we heard the employees, and we decided to set a dollar amount, which is not a big dollar amount, but we set a dollar. You basically say, if your grant is all this amount, then you are not -- your vesting will be immediate.

So, because of that change, we are now raising the expense for Q3 this year for RSU grants because of our reason, meaning you will be more than the one quarter that we have guided previously. Certainly, the other side of the coin is that when we increase the expense this quarter, the following quarters, the expense will be less. But that's what we are trying to do here.

So now the first quarter will be roughly about 50% of the total grant? Will that be changed again in the future?

There's a lot of factors affecting the vesting portions. We expect around 40 to 50 around of the total grant value will be vest on the grant date, while the balance will be vesting equally over three installments in the next three years. I think this is to stay competitive in attract and retain talent, and we're competing with our local peers with the same talent pool. So we revised our RSU programs vesting schedule in a way that junior engineers, who are granted rather small amount of shares, will have a large portion of their granted shares vest immediately on the grant date.

I think this is a practice commonly seen in our Taiwan local competitors. But as we grant RSU only once a year, scheduled in end of September, which are a significant portion we will vest on the grant date. So we will see a bump in the share-based compensation in September of each year and a low GAAP EPS in the third quarter as well.

Will this ratio stay the same for 2007, or will it not be changed?

2007 -- probably, I think the rule will stay constant, I think. That's in response to the junior or new engineers, meaning the rule implies two things -- one, [30%] of total profit of that year. I really don't see that changing. Certainly, anything can change. But at this point, that has been our rule. The second part is the vesting schedule. We are sticking to the full four-quarter installments vesting schedule for people such as myself. So this new rule doesn't benefit myself, for example, or more senior people. However, for people with very little grants, they do benefit. So every year, the portion of the spend -- in other words, total grants in that very quarter when the grant occurs will vary slightly, depending on the distribution of that shares to these employees. But, based on our HR's estimate, as Max said, you'll be 40% plus in this current quarter.

Another clarification on this is that regarding the third quarter value of roughly about $11-$13 million total, can you explain to us from a share perspective in a way that if the stock price is quite deviated from where it is today, will this amount be changed?

If the stock price is at today's level, the grant amount will be approximately 2% or slightly less than 2% dilution of our total outstanding shares, which is in line with most of the Taiwanese peers.

(indiscernible) I probably understand your question. (indiscernible) if the variation multiples of [starters] surge or decline potentially from [place] levels, how are we going to make the adjustment? I guess that is your question.

I think, in theory, when for example -- I will take a very extreme example -- P/E multiples all of a sudden become 30 times for some reason, in theory, we are likely to grant less shares while offering slightly more value than 30%. I think that is likely. However, on the other hand, if the multiples become very, very low, (indiscernible), then there is a possibility that we may reduce the 30% number in terms of value. However, we may have to stop us from a bit more share dilution because we will need to gross a bit more for our employees in terms of value -- in terms of shares. Does that answer your question?

Yes. In that case, is there a cap (indiscernible) to know how many shares or it is a value that you will cap in those two cases, not only--?

Last year, we had a shareholders meeting, which granted the Board of Directors a total of 10% over five-years period. We have seen that as our cap at the moment. So last year, we granted 1.7%. This year, if I recall correctly based on such [graduation], you will be about 2-ish%.

Are you considering some stock buyback plan?

(multiple speakers) We have not formally announced it. So I really can't say that we are considering it. But it's certainly an obvious subject we must consider.

Actually, during the past quarter, we had quite a few discussions with investors and that provides a lot of advice to Himax. So we are seriously considering all their proposals and (indiscernible) advices in any about the Company's dividend policies or any other (multiple speakers) --

Or share buyback programs.

Adele Mao, Susquehanna Financial Group.

Can you share with us your view on the near-term growth opportunity you foresee in Mainland China in terms of unit growth with your existing customer or discussions with any potential new customer?

In Mainland China, there are really only three major panel players to date -- i.e., SVA-NEC and BOE [High Disk] and other one is called InfoVision. InfoVision is still in its very early stages of mass production, so any volume growth from there will have to be for the future for anybody. We have been a major supplier to SVA-NEC to the extent that market share-wise, vis-a-vis this particular card, I think it is now becoming difficult to grow because it's very much the feeling already.

We have seen our volume growth coming from this particular [calling] and this is expected to continue in Q3 because the customer is now getting back to full production following certain slowing down. Because at the time, they were trying to increase their capacities by using the same premises to try to increase the capacity. So the production got somehow disrupted at the time.

So they now appear to be of full speed again so that is good. And also, this is after they are having finished their capacity increased (indiscernible). So that, if you like, is China-domiciled companies are our customers. But if you think about shipments to China -- and certainly as you know at this stage, everybody has their module operation based in China so that will apply to everybody -- so I think I will focus my answer to companies who are headquartered in China.

Are you still seeing Chinese market more focused on smaller 15-inch, 17-inch displays? Or are there potential opportunities to get into larger segments?

We do have design process into larger segment areas, including TVs. But in terms of real volume, I think you will come -- [see primarily flow] kick in -- 17-inch monitors, as you said correctly.

[Tony Sandhu], HSBC.

Great numbers in a tough quarter. I just have a couple of questions here. I may have missed earlier comments from you. But if I go back to second quarter and talk about unit growth in the PC side and in the television side and probably also discuss the ASPs in both categories if you could.

Right. We suffer from the most on the TV side because I think in the (indiscernible) on a worldwide basis, TV demand was very, very slow, and there were a lot of reasons for disappointment throughout the whole wacky environment. So we are not exempt from that. So TV used to account for about 30% of our total business. In Q2, it accounted for about 20%. That has been reversed by OA, meaning monitor in notebook.

We are indeed seeing TV demand coming back in Q3. Although, I won't apply this comment to every single customer. But certainly, certain customers are still adjusting their inventory levels. As I said earlier, it could be a case where their inventory level is just higher than normal. It could also be a case where they are trying to reduce the so-called normal inventory levels, meaning they are trying to write a tighter operation to save cost. So I think we are seeing both happening in today's market across different customers. That is why we're not overly bullish on -- that is part of the reasons why we're not overly bullish on Q3 operation or Q3 demand.

I think there is no clear-cut, like before June 30 -- that time, after June 30th, that's good times. So we think that demand picking up is gradually. So as some of the customers or some models are still experiencing inventory corrections, as Jordan just mentioned, in the early part of the third quarter. But overall, the demand we're picking up throughout the entire second half of the year, there is no clear-cut -- say, okay, second quarter versus third quarter per se.

On the TV unit side, second quarter over first quarter kind of in sequential terms, were the units flat or actually the units were down like 10% or 20%?

The TV revenues actually accounted for more than 25% -- well over 25% in our first-quarter revenues, as customers experienced a very strong demand in the second quarter as a result of World Cup. But in the second quarter when the results came out not as good as expected, and people slowed down the procurement of TV drivers. So in terms of unit shipments, we will be talking about around 15% sequential growth for TV.

That's in the third quarter.

In the third quarter.

Was that flat in second quarter, or was it down? I guess I'm trying to get to the units here. I know you have made comments on the revenues.

You are certain it's down, down substantially, down by about 15% (multiple speakers) of the third quarter.

I know you probably don't want to comment on specific customers. But in terms of you said Chi Mei is now down to about -- in the second quarter was down to about 50% of revenues, and that was largely on the television side I assume. So in the television side, how big of a customer are they? Obviously, they are probably much bigger than 60%-70%, but can you give us an idea -- the non Chi Mei customers, what is the percentage there in the second quarter just for the television side?

We have two major TV customers so far for (indiscernible) IC, being Chi Mei and Samsung. As you know, they are only about five serious players in the market today for TV shipment. We do have a few other customers having TV [dugger] shipments. But in terms of size, they are probably very small compared to the top five. So for Himax, I will say it is primarily Chi Mei and Samsung. I would say in terms of threat balance, Chi Mei's (indiscernible) is much bigger than Samsung, given that the fact that the size of the two customer shipments at similar levels. Certainly, Samsung is probably bigger than Chi Mei, but they are not substantially different. But we account for a much bigger percentage from Chi Mei.

One of your major competitors in Taiwan in terms of unit shipments, they saw a very sharp drop-off in June. Basically, April was okay and then May dropped a bit and then June was just off the cliff. Is that something -- and then July has come back strongly for them. Is that the kind of pattern you saw, or was it more kind of a -- was it a little bit more stable than that?

A bit more stable than that but it's also tightened in the sense of the direction. In terms of magnitude, we're more stable than that. June typically, as you know, it is end of the quarter. In fact, even worse, the end of the second half, so people in Taiwan are required to prepare full prolonged financial statements. So that sort of explains.

Just talking about the product mix going forward, you expect something that TVs will potentially be -- TVs would potentially grow at 15% unit terms? How confident are you of this, or is that based on kind of order flow that is currently on-hand? Can you talk a little bit about that?

Quite confident. We have actually -- I have actually personally shared with -- you know I just mentioned there are only two, right (multiple speakers) --

Right.

-- first customers. But they are quite confident about it. And (multiple speakers) system potential upsides, I wouldn't count on them. Anything could happen in this environment today.

I just want to talk about the prices as well now; we talked about the units. Was the pricing pressure more on the television side than on the OA side in the second quarter?

I think similar. I think the customers in second quarter are all going sort of like upside down. So customers I see very aggressive in terms of pushing for lower prices. So we did experience very high-priced pressure across the board in different product areas. I commented earlier, we did lower our prices more aggressively in gate ICs, meaning IC with COF packages. Because that is an area where the price declined more than other parts of our components. So it is a reflection of our costs, so we were able to push down the price more aggressively compared to other product areas. We are seeing similar things in Q3, although probably not to the same extent. We are also seeing price being virtually stabilized, not to say that there will be no further price pressure in Q3 or Q4, but certainly not to the same degree as we saw in Q2 or even Q1.

On the financials, Max, could you talk a little bit about this ForEx gains here in the statement?

Yes. We paid some costs in NT dollars. And when the [India] dollars depreciated, we're paying in weak currency, so we'll have some foreign exchange gains.

Let's see how it is going to be like going forward, but we are now changing our trading currency with all our suppliers to US dollars. There used to be some yens and some NTs. Now it's all US dollars predominantly. Our sales have been denominated in US dollars from day one. So we're hoping to reduce the pressure of foreign exchange gain or loss fluctuation going forward. Let's see how that is going to resolve. But that is what we are trying to do now.

Glad you didn't do it in the second quarter.

Dan Scovel, Tokeneke Research.

You mentioned at the beginning of your prepared remarks the percentage difference in revenue between related and non-related parties. I am sorry; could you repeat that, please?

Right. I said revenues from related parties declined 21% after a strong quarter in first quarter. Wire revenues from unrelated parties increased 29%.

Was down 2%-4% and up 29%?

No. I repeat -- our related parties up 29%, related parties down 21%.

What is the number of customers you have as a non-related party?

Many. I lost my count. Many, many, many. But certainly, having said that though, we are still a company with a lot of customer concentration. So you talk about -- let me see. You talk about revenue of $1 million or above on a quarterly basis then we have around close to 10 such customers. But certainly, there has been a lot of concentration on a few accounts including CMO and on (indiscernible) CVT in this space, SVA-NEC, Samsung. Now we have [HeadStar] in the last etc., etc., large panel customers. Each one is quite big for us.

Eric Lin, UBS.

You mentioned about the guidance for third quarter, which you said the revenue up low teens -- sorry, revenue up mid teens where Chi Mei up low teens. So is that implying the blended ASP you're seeing some uptrend in third quarter?

No, no, no, no, no. Let me clarify that. We said our revenue top line was up by mid-to-high single digits. And units are likely to be up by low teens, meaning there will be around let's say, a 3% price erosion during the quarter.

That's when ASP was down around 3%, right?

Yes. So let's say most of our major large panel customers, their main stream has been what we call high channel count [struggle] ICs already. So you have liked, when you talk about units versus revenue or ASPs, sometimes it is a confusing number because you're really talking about different specs for each IC. So how do you compare ASP?

But this phase has been a situation where most customers have adopted high-channel card ICs. So now, when you talk about quarter over quarter, ASP change, it is now more meaningful, and here we are expecting around probably 3%.

That's actually my next question. The high-channel server ICs, can you give us a breakdown of the moves right now compared to a typical 380 or 480 channel? What is the progress of the 600 channel or even 700 channel business?

Okay, I can't comment on specific customers. But I will just say that with very few exceptions, we are seeing customers already adopted high-channel card, meaning 642 channel source of ICs for their monitors, not for projects already.

My next question -- actually, it's packaging. Can you give us some data points about the breakdown in terms of packaging within your driver ISC sales?

You mean the costs?

No, no [brit] sales (multiple speakers) --

I mean the packing-type costs and COG event.

TCP is very small; it's only used for certain TV projects, not all TV projects. So it's quite small. I would say TCP would be all together probably (indiscernible) COG. COG in our case for large panel, COG is applied to get drivers only, and not every customer gets driver. But certain customers get drivers. They have adopted COG solution for their gate drivers required for source driver; they are still using COFs.

So COG in our case is primarily in the case of large panels for gate drivers. While it's literally all of our small and medium-sized panel drivers and mobile phone panel drivers, they are all of COG packaging. So in terms of units, I would say around two-thirds.

George Chang.

Just very quick, your revenues from Chi Mei was down 21%, more than 20% in second quarter. That's in line with the production cuts we have seen or heard from Taiwan. But your revenues from other clients have increased huge, like more than 30%, during the second quarter. It sounds like it's more monitor driven. Could you please elaborate a low bit which customers you have been gaining market shares from? If not the names, perhaps which countries. Was that Korea or China or any other applications other than monitors?

The follow-up question would be, it seems like your margin decline during the second quarter has been a little bit greater than your competitor. So does that imply that you have been gaining market share at the price of lower margins?

Right. I will address your last issue first. The answer is, I don't think so. It doesn't really work this way. The design in lead-time is very, very long. So let's say last year -- second half of last year, we didn't know exactly what the number was going to be until this year. We did know that we are likely to gain share out of certain accounts because of the design schedule.

What we weren't sure about was -- first, the price at the time of the production and secondly, the customer's total output, the product mix and so on. So first details we didn't know. However, we did know well in advance which customer or which customers were likely to come in during which time. We did foresee new additions coming in around Q2 of this year. Certainly, such customers were then in Q2, entering to mass production in certain of their new projects, for example, and they came to us and they tried to bargain for a very aggressive price. So our price discussion is an ongoing thing, while design is a very, very long lead-time kind of process.

As far as which customers we have been gaining market share, I would just name a few. They include Samsung, SVA-NEC, (indiscernible). Certainly, in the case of Samsung, it's primarily the TV and for the others certainly monitors.

So you are saying that a lot of second-quarter improvement from other customers have been kind of anticipated over the past few quarters. If that's the case, do you anticipate similar patterns going into second half, i.e., are there going to be much new projects coming from these customers? Or is that third-quarter revenues growth is more going to be driven by Chi Mei's rebound?

I would say -- put it this way. Firstly, we are still feeling quite good about our long-term progress vis-a-vis such new customers. The reason I say that is because we just look at the design projects. The quicker they are going forward, they will be relying more upon Himax or less upon Himax. I would say the overall trend is that customers are relying more upon Himax than before.

So we have our existing customers, existing major customers, and we are not seeing our market share position or our perception in the minds of their engineers or their top managements. We are not seeing this deteriorating. But in many cases, we are seeing this being enhanced. So that is why we are still feeling quite good about our long-term prospect in terms of market share.

Having said that though, there could be as I said earlier, a very good example whereas there is one of our major customers, one of their major product areas before which were suffering from losses -- heavy losses. They had decided that they should switch their production line into other product segments. So, although they did not do that intentionally to, if you like, penalize Himax for full service or whatever, our volume suffers because we happen to be the sole source provider in that particular project.

So such (indiscernible) is possible. I would just say that if I knew across all this OA new customers, I think we're being more and more relied upon by such customers going forward.

Right, understood. I was just trying to -- one, figure out whether your third-quarter revenue guidance would be more driven by Chi Mei or the third-party customers. Quite simply, you aren't guiding (multiple speakers) --

By Chi Mei.

More by Chi Mei?

Yes.

Tony Sandhu, HSBC.

In response to an earlier question, you mentioned that most of your customers have already adapted the higher capacity channel ICs for OA applications. Can you put some more color on as a percentage of your output, how much of it is already 640-channel and above?

This is a rather tough question because for example, certain TV projects could be very suitable for 480 channel, for example. So I'm not going to provide a lower channel compared to 482. So I guess I will give you a guesstimate. The definition of which is that for the typical 17-inch or 19-inch monitor, here, you are really talking about a choice in between [300 of 84] channel, the history, or 642 channel, the present.

If this is the question, then I would say three quarters or above -- yes, three quarters or above already in high channel, meaning 642. Again, I can't name names. But many of our customers is 100% already or very close to 100% but with a few other customers that have close to 0. But they do have new projects for 684 channels. But overall basis, I would say it's three quarters. But that doesn't provide cost (indiscernible) for the customer obviously.

On the television side, obviously this is much lower, as you were saying earlier.

Yes.

Gentlemen, there are no further questions.

Well, thank you, everyone, for joining us for today's call. I will say that concludes our investor conference today. And if you have any further questions, please do feel free to call us directly. Thank you.

Ladies and gentlemen, this concludes today's teleconference. You may disconnect your lines at this time.