Cognex Corp (CGNX) 2005 Q3 法說會逐字稿

Good morning, ladies and gentlemen. My name is Ian and I'll be your conference facilitator today. At this time, I would like to welcome everyone to the Cognex Corporation third-quarter 2005 earnings conference call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question-and-answer period. (OPERATOR INSTRUCTIONS). It is now my pleasure to turn the floor over to your host, Mr. Richard Morin. Sir, you may begin your conference.

Thank you and good morning, everyone. Last night we issued a press release announcing Cognex's earnings for the third quarter of 2005. For those of you who have not yet seen this report, there are copies available on our website at www.cognex.com. The press release contains detailed information about our financial results and because of that we are not going to repeat most of that material. I would like to emphasize that any forward-looking statements we made in the press release or any that we may make during this call are based upon information that we believe to be true as of today. Things often change and actual results may differ materially from those projected or anticipated. You should refer to the company's SEC filings, including our most recent Form 10-K for a detailed list of these risk factors. And now I will turn the call over to Bob Shillman.

Thank you, Dick and good evening, everyone. I said good evening because I am currently speaking to you from Tokyo where it is 10 PM. I am pleased to report to you on third quarter of 2005. It was a very good quarter for Cognex; both revenue and profits increased significantly on a sequential basis and net income as a percentage of revenue was 19%, which is just a tad below our long-term target of 20%. Although I won't be satisfied until we're over the 20% mark, the level of profitability that we achieved in this Q3 was still quite an accomplishment.

By the way, I do want to point out that the earnings report issued last night reflects the full quarter's expensing of the excess purchase price from the DVT acquisition, which closed in May of 2005. Excluding the amortization of acquisition related costs, operating income for the third quarter would have been 15.1 million or 26% of revenue and net income would have been 11.7 million or 20% of revenue.

In Q3 we saw a nice pickup in business with our customers in the semiconductor and electronics capital equipment markets. Orders from the segment increased 10% on a sequential basis and revenue was up 27% due primarily to higher level demand from customers who make equipment for the back end of the semiconductor industry. That's primarily wafer probers and die and wire bonders. However, looking at the year-on-year comparables, revenue from these class of customers in the third quarter of 2005 was far below what we have reported for the same period of the prior year. As a result of that and as a result of business increasing elsewhere, that segment represented a much smaller percent of our company's business than it was a year ago.

Decline year-on-year and revenue from semiconductor and electronic capital equipment market was completely made up for by increases in the surface inspection and factory automation markets, both of which reported record revenue for the third quarter. The factory automation market, even though orders were soft as usual during the summer months, we benefited from the incremental revenues of DVT.

Speaking of DVT, the integration of DVT into our operations is essentially complete and I am pleased to report that we have already seen some early indications of its ultimate success. Revenue from DVT products was at a record level in Q3 and was at its highest level ever on a monthly basis in September.

In addition, we realized synergies from that acquisition sooner than expected and had savings in areas, including marketing communications.

So in summary, business has increased quarter-to-quarter and we expect that trend to continue to next quarter. Revenue for Q4 '05 is expected to increase by 5 to 10% and will result in revenues somewhere between 61 and 64 million. And at that revenue level earnings are expected to increase somewhere between $0.24 and $0.27 a share to end up at $0.24 and $0.27 a share.

Now, operator, we're going to open up the conference call for any questions that you have for the Cognex team on this side and we're prepared to take the first question.

(OPERATOR INSTRUCTIONS). Alex Paris, Barrington Associates.

Just a quick question or a couple of questions maybe on DVT. It has been better than expected for you at least both in the second quarter and the third quarter and is it mostly because are the sales running better than expected or are you finding it more profitable than you thought or are you just -- the integration just went faster?

No. I think it mainly comes down to the revenue being higher than we expected.

We have also been able to realize some of the savings a little bit from the integration and synergies of the two organizations. Some of the savings have come a little bit quicker than we had originally anticipated, Alex.

And was there accretion? You had $0.01 I think in the second quarter.

That's right.

It is hard to measure that now, Alex, because of what we have done with the two organizations. Some of the marketing that used to be done at DVT is now all done up at Natick. There has been a co-mingling of the sales forces. So it's very difficult to try and say what the bottom line impact is of DVT right now because through the DVT channel they are also selling some of the In-Sight and Checker products. So it is almost impossible to get down to a bottom line number as to what the impact of DVT was.

Are you seeing an increase in In-Sight and Checker because of the new distributors that you brought onboard?

We have seen a little bit but this past quarter was a slow quarter for the factory automation in that during the summer months, historically, has been very slow for that segment of the business. We saw a nice pickup come September after Labor Day both in the U.S. and internationally. But it really wasn't a good quarter to be able to see that there was a major increase coming out of those distributors.

Just one other quick question. Just looking at the DVD and the markets that they are dealing with that compared to what you were dealing with in your factory automation business, do they bring any new substantially better penetration of an industry from what you had before them or are they pretty much mirroring the industries that you are doing well with In-Sight and Checker and so forth?

No, Alex, pretty much the same industries; automotive and automotive suppliers were big for both of us. I think they did a little better than us in the packaging end of the business perhaps in pharmaceuticals. Do you have any more data on that, Dick?

No, I think that's pretty accurate. They are into an awful lot of the same industries that we were but they definitely were a little bit ahead of us in packaging in pharma.

Suresh Balaraman, ThinkEquity.

Dr. Bob, in terms of your Q4 guidance, is there seasonal or positive lumpiness that are happening that may not happen in Q1 of next year? Is there any reason to expect that economics continue to grow at a modest rate into Q1 and Q2 of next year?

Well I think the biggest uncertainty is the continuation of the rather robust orders we are now getting from semiconductor and electronics. We believe the underlying growth that we're going to get in the factory automation should continue. We think that should continue for the next few years, as a matter of fact. Surface inspection -- the growth in surface inspection was unusual this year so far and may slow down. I hope not but it's hard to believe that the growth rate will be sustained. The biggest wild card is the semiconductor OEM.

Now that you're in Japan, do you get a better sense of how some of your customers (indiscernible) thinking of the near-term also beyond --?

I haven't yet had any customer visits so I can't comment on that but I -- I can't comment from the standpoint of my visits that I haven't had but I can tell you that bookings are very strong from the back end. We're seeing very nice orders and the orders are being pulled in rather than being pushed out.

Any (indiscernible) on the Checker?

Checker continues at approximately a $3 million run rate right now. The positive news in addition to that -- although it's not a $10 million run rate is that it is growing quarter-to-quarter and also we now have a better understanding of why it is taking longer to grow than we had anticipated. The reason appears to be that we are selling into a –- it's not what we first thought that it was the way we were distributing it that we had the wrong distributors. That is not the case. It appears that the customers that we're selling to are very different than the ones we have sold vision systems to in the past. These are customers who are used to buying sensors and for them it is a dramatic change to go from sensors to a vision system, our expert sensors I should say. Although our expert sensors are far easier to use than vision systems and they are far less expensive than vision systems, they are more complex than sensors -- than photosensors and they are more expensive than photosensors. So there is going to be a learning curve here. We hope it will be shorter than the learning curve of vision was in the past. We expect that to be the case but that is the reason why we are only growing at the rate we are growing and not three times faster.

Jed Dorsheimer, Adams Harkness.

I guess perhaps a follow-up to the first question that was asked. Dr. Bob, could you give us an update on how the product line reconciliations have gone between Cognex and DVT and what measures you have done so far, just an overall update there?

The initial thoughts were that it would be rather simple to just eliminate certain DVT products and just replace them with Cognex products. And that turned out when we looked at it -- first of all, it didn't necessarily make sense to do because DVT's margins frankly were better than ours. Their margins on the DVT line was like 80% and our margins on In-Sight are more like 77%. It's not a big difference but why discontinue a product line if it is more profitable than our existing line.

We also found that customers were rather happy with the DVT products because in particular at the lower end of the line their product is easier to use. Less capable, but capable enough for many problems that customers were solving, which is why it was a good company, a great company and that is why we bought it. So we have not done much consolidation as of yet and I think only minimal consolidation is planned in the future. We expect that ultimately -- and ultimately in maybe 8 to 12 months that the high-end line will be merged into or discontinued and it will be called In-Sight and the lower end of the line, which is superior to ours and easier to use, will continue and be called DVT.

We do want to eliminate confusion with the distributors and with the salespeople and with customers. We don't ever want to show them equivalent products and have them choose between the two virtually equivalent products. That doesn't make sense. But we're very pleased with the engineering team down there with what they have done and the manufacturing team and we're just going to focus them on ease-of-use issues and continue toward less expensive products and that is what we're going to be doing.

And then at our conference you had mentioned a couple of potential -- that there were a couple of acquisitions; one large, one small I think is how it was outlined. I was wondering if there is any update now that it looks like cash is at a pretty healthy level and I just wanted to get any update if we should be expecting something in the near-term here from you guys?

I would say no and I can expand a little bit more on that. The one large thing that we are looking at we decided that we are going to try to do that ourselves rather than go after that company. Rather expensive and complex and we are going to try to design that into the product ourselves.

The smaller company is something still in our sites but we have reviewed the contract with that company and there are certain aspects of the contract that are already in existence that we're not crazy about. So nothing is going to happen, I would say. I don't see anything closing this year. Probably even in the first quarter, I don't see anything closing unfortunately. We would love to be more acquisitive. With Jim on board -- we have -- and his team, we have the management talent to be able to manage more acquisitions and integrate them. We have a pretty good record -- a very good record of that I would say. We have just got to find more fish and put them on the boat. It is not easy finding acquisitions that meet all of our criteria.

Last question, could you talk a little bit and maybe Dick wants to take this one or you, Dr. Bob. If you could talk a bit about -- as we look at the guidance here, it looks like OpEx is expanding the midpoint of the numbers at the same rate as revenues. I was wondering if you could just maybe give a little bit of color on why OpEx is growing into Q4 given the guidance.

I think we had some savings in Q3 that came a little bit earlier than expected. But some of that in Q4 is just getting over now that we have had an opportunity of the two organizations getting together, looking at its sales and marketing programs, it will be an opportunity to kick some of those off in Q4. We have got some sales meetings that are planned, kickoff meetings to get the folks in line and trained and see everything that is going to be happening in '06. So that is part of those increases that we expect to incur in Q4.

Richard Eastman, Robert Baird.

Could you break down the book-to-bill? There was some reference in the press release that the book-to-bill might have been perhaps just under one. Could you break it down by the three business groups: the factory automation, semi, and SISD just to give us a flavor for how factory automation and semi are?

Okay. I think -- it's going to take a couple of seconds as I go flipping through here.

Maybe I'll just jump ahead here. Dr. Bob, if I try to adjust for DVT's revenue, it would appear that the factory automation business year-over-year was up maybe in the midteens. I don't know if that is the right math but if it is it would suggest a little bit below trend growth in the third quarter. Any feel for that?

That would surprise me because historically the factory floor has been growing low 20s but it could be an anomaly. I didn't look at those numbers. And Dick, we don't get book-to-bill every quarter, do we?

Yes, we do. We disclosed what the total book-to-bill was but we don't give the dollar amounts of the bookings.

My guess is that the SISD book-to-bill is well below one because that business has been so strong and what I'm trying to get at is was the factory automation book-to-bill and the semi book-to-bill well over one? Is that kind of how the mix would look?

The factory automation book-to-bill was right around one. Semi was -- let's see -- semi was right around one as well and SISD was more like 0.96, so whatever.

So that actually held up pretty good then (indiscernible) that business has been tracking, SISD orders.

And actually the factory automation being that high was pretty good as well because normally we always have -- the summer months or the third quarter has always been a slow quarter on that part of the business because so many of the operations shutdown for vacations and we did see a strong rebound in the month of September. So that was pretty good on the factory automation side.

Dr. Bob, your take on the semiconductor business, it sounds like things continue to strengthen there. There is a feeling of thought that we're still seeing seasonal strength on the back end. Is there any message in your order flow? Are you seeing production orders or anything that might suggest that the cycle might be a bit longer in nature, longer-term nature?

You could always look at the tea leaves and read that into it. But I would rather -- we don't try to draw conclusions because it doesn't change the way we run our business. It makes sense to look at data if indeed conclusions from that data would change your actions. But we don't run our OEM business that way at all. The factory floor perhaps we changed some things around with regard to number of salesmen that we'd hire and marketing programs and trade shows but the OEM business, if it comes in, we build it and ship it. If it doesn't come, we don't build it and ship it. So it is sensible for you to ask these questions. It may have some bearing on other companies and other investments but we just don't draw the conclusion and we don't care.

It would be great if it continues but it doesn't make -- if it happens, it happens. The results -- if it continues longer, we will be profitable longer. If it doesn't, then that segment of the business won't.

There was certainly a comment in the spring that inventories at the OE price had stretched out. Have we at least seen those shrink a bit?

They are not increasing and they have -- to the extent that they have shrunk, it is very minor. There hasn't been a major movement in their inventory levels for the most part.

Antonio Antezano, Bear Stearns.

Good morning.

Hi, Antonio. This is your first conference call.

That's right. Thank you.

Welcome.

Thank you. I just had a follow-up question on DVT and the seasonality. What is the normal seasonality between Q2 and Q3 for the factory automation business because if we try to exclude DVT from the third quarter and the second quarter, it seems that there was a decline there for base revenue? So I was just trying to understand how the seasonality plays out from Q2 to Q3. What is normal seasonality?

If you go back and we have got charts that have gone back like the last four or five years, typically we always see a decline in the factory automation business in July and August and that is principally due to vacations, plant closings around the world. In Europe, they shutdown for a month at a time. Here in the U.S., it is a week or two or whatever. But consistently, if you go back the last four or five years, we will go from a high level of bookings in June and we will see declines of 20 to 25% or so month (indiscernible) as we get into July and August. And then historically, and it happened again this year, we tend to see a big pickup in September when for the most part all of the factories are back in production. So it is a normal trend that we see every year in the third quarter and it was no different this year.

Alright. And then in terms of the target revenue for Checker, before I heard that there was this ball (ph) of 10 million for the first year. Is that something that still holds?

Well what would make you believe it even if we told you that? Our track record at predicting what Checker will do is obviously pretty bad. But I would say what we expect to happen is currently would be double this year's rate -- this year's results and we will probably end this year somewhere a little short of $3 million I would think.

Actually, Bob, I think for the year, we should be somewhere between -- this year, we should end up more between 3 to 4 million.

Okay. I thought the last time I -- so we're talking somewhere around 6 to 7 million.

Doubling that next year maybe, yes.

And then one final question, regarding the cash level, you haven't done, I believe, any share repurchase since 2002.

That's right.

Any plans to do that? It seems that when you have a very high cash level, you contact the share repurchase the following year. So should we expect something in the near-term?

No.

Okay.

We prefer to use the cash for acquisition but as I mentioned earlier, it is hard to find the acquisitions and then the new accounting rules make it very difficult for acquisitions to be justified. This is a similar problem that we're going to be facing with stock options I think and I don't understand why the government wants to continuously add the burden to companies without any benefit to society. I think it's a real problem. The most recent legislation is whether it is Sarbanes-Oxley or stock option expenses are simply things that make America less competitive in the world and that is a bad thing. I just don't get it.

Jim Ricchiuti, Needham & Co.

Dick, a question for you. I want to understand something a little better going back to the seasonality of the factory automation. If you look at last year's Q3, what kind of sequential growth did you see in factory automation? I don't know if you have that data in front of you.

Sure we do. What is the question again?

Sequential growth, Q2 to Q3 factory automation.

Factory automation, Q2 to Q3, last year?

Yes.

Okay. There was a decline last year.

Okay. There was a decline last year.

Absolutely.

Okay. That's what I wanted to just clarify. How big a decline was it?

Close to 10%.

And then just to switch gears on the surface inspection of the SISD business, it continues to be stronger than I would have thought and I was just wondering if you could elaborate on where the strength is coming from, particularly vertical markets and where you have -- I have a follow-up to that. Why don't you go ahead with that one?

Well, we are all in awe of the great job that Markku Jaaskelainen who has been the general manager of that division for the past three years I think and what he has done is remarkable. We wouldn't have predicted that you could grow surface inspection as fast or to the level that he has grown it.

What we are doing now is making headway in the steel industry which is where we fell back for a couple -- quite a few years matter of fact when we focused on paper. So that was his plan. The paper industry is far larger but now we are winning orders in the steel business and I don't think we have issued a press release about it yet but we have just gotten a very nice order, one that -- our first one from a major steel company in the Far East that we will announce soon and we expect three follow-ons from that same country. So we continue to beat the competition. I don't believe that the market is growing as fast as we are growing or matter of fact we know that the market is not growing as fast as we are growing. We are just beating the heck out of the competitors.

And Bob, the order that you refer to was an order that is in the bookings for the SISD business?

Did it come in the September or October, Dick?

It came in, I think it came in in September.

And in the past you have also talked about some efforts to improve the gross margin for the SISD business --

That's right.

I was wondering if you could give us an update on how that is going.

Too slowly for my liking. So the main thing is to switch gears. We were on a path to win business, not at any cost. You can see our margins are decent but I will tell you that I think the operating margins -- can we talk about this, Dick?

Yes, in gross terms.

The operation margins of SISD turned out to be like 20%, 20% in Q3 which was remarkable for a capital equipment business like that. But it wasn't because of an increase in the gross margin. It was an increase just because the revenue came in higher for the same operating expenses. So it takes a while for the gross margin to increase. Only two ways of doing it, increasing price or decreasing product cost. And we are going to see if we can work on both of those. Markku has been instructed and he's motivated and bonused to get that gross margin higher.

There was an improvement, a slight improvement in the margin, gross margin quarter-on-quarter and the main thing helping was the fact that expenses, operating expenses were essentially flat.

You said gross margin increasing, Dick.

Yes. The gross margin percentage increased, did increase during the quarter. Secondly, expenses -- I'm sorry?

Oh, secondly, yes.

And then secondly expenses were flat, so with the increased revenue level that really helped the operating income line.

That's right.

I just had one follow-up on the Semi and Electronic Cap equipment business, any sense as to how the bookings have trended through the month of September? It's still early in October but I'm just curious --.

No, I can tell you that October looks very good. Some very big orders have come in. But I didn't understand the question other than that.

That was it. Just curious how the bookings were.

It's continuing.

Tim Culvey (ph), Fortich (ph) Capital.

I guess this is a question for Dr. Bob. I'm looking at the revenue by geography, and I see 2004 looked like it was pretty strong for Japan. Since you are sitting there, I guess you can give us some color on what Japan is like, their global economy or their regional economy how it is improving and some color on why Cognex is kind of falling behind there.

I think you are new to the company and you may not be aware that in 2004, we had a substantial pickup from the semiconductor and electronic OEMs who are mainly in Japan. So that is why you see the nine-month to nine-month decline and also the year-on-year decline from 41% to 30 quarter-to-quarter. That has only to do with the pickup or slowdown in the semiconductor industry, which occurred in 2004.

Well, can you give us some color on how Japan looks just (multiple speakers) --?

I want to make it clear, we haven't lost customers. Nothing has happened to lower that percentage other than the fact that those customers and their competitors, that whole industry, is doing less well than it did in '04. Go ahead.

Thank you.

(OPERATOR INSTRUCTIONS). Alex Paris, Barrington Associates.

Just from my edification, just looking at photosensors versus your expert sensors, what is the significant difference in the technology? Is it photosensor, is the technology on the way out or is --?

Good question, Alex. Good question. A photosensor is a single -- you can view it as a single pixel vision system. It just has one sensitive area to light. That is it. And it reports either that light -- sufficient light came in or sufficient light didn't come in. That is all. It's just on or off. Now you can do a certain amount of things with a photosensor and people have tried to do those things. But Checker on the other hand has 30,000 -- let me see which version I'm talking about -- thousands, tens of thousands of pixels and you can do a lot more with tens of thousands of pixels than you can do with a single pixel.

Is this specifically machine vision or not? You don't -- obviously, it is just one little sensor. There's not a computer -- there's vision software in it?

Of course. Yes, it is a vision system. However, we are marketing it and packaging it as -- we call it an expert sensor, because the people who we are selling it to would be intimidated if they knew how much technology was inside that little box.

Now, if Checker were wildly successful, which industry would you see the biggest jump in? Where is this low cost --?

No, discrete manufacturing anywhere. It would not matter, Alex. It is not industry specific. It is not at all industry specific. Discrete manufacturing, anywhere where items are made that you can pick up or that are discrete, separate from one another, whether it's pens or pencils or motors or disk brakes or chocolates; wherever the product can apply to all those products. Now having said that, most discrete manufacturing in the world is involved in parts for cars, automobiles. So most discrete manufacturing is automobiles. That's the largest segment. So probably not 50% of the entire world market, because there's manufacturing toothbrushes and there's manufacturing pens and everything else that you buy today. And if you added those up, it's probably far more than automotive. But the single largest segment of manufacturing discrete items is probably for cars.

And would that imply that you would be selling more to parts suppliers as opposed to --?

That's correct, and that is -- currently automotive does represent our largest segment for In-Sight. I don't know if we've done the segment analysis for Checker, but In-Sight, the largest customers are -- the largest segment are automotive suppliers.

Okay, that's very helpful. Thank you. One other thing, talking about tea leaves. Is there -- do you see any correlation when you see the book-to-bill ratio for the North American semiconductor equipment -- I know you've got more of your business is (indiscernible).

Yes, that is something that we look at and that does predict the growth of our business. Actually, it's not the North American semi that we look at so much, but it's the Japanese ratio that we look at. That's right.

And is there a correlation between the North American and the Japanese one that you see?

I would think so, but I don't recall. Maybe Dick can answer that, but I know there is a high correlation between our bookings and the sale of capital equipment in Japan, which is readily available.

I ask because there was a very big jump in August to 105 book to bill in the North American equipment.

You know, I just don't know but maybe not because the North American -- I'm not an expert on this, I used to be -- North America has more front-end equipment than back-end equipment and front-end equipment doesn't have as much vision on it as back-end equipment. So it could be that we are not as strongly correlated with North American semi bookings. Do you have any comments on that, Dick?

That's pretty true. In North America, most of our semiconductor capital equipment OEMs tend to be in the front-end and we have not yet seen the kind of pickup that we've experienced from some of the back-end customers that we've seen in Japan.

Another reason that would be, by the way, is that front-end tend to have much higher costs and much larger systems and so when someone spends $5 million on the front-end, it could only be two vision systems that are in there. Whereas, if you spend $5 million on wire-bonders, you're talking about hundreds of wire-bonders.

That's helpful. Just one other quick question. What is your average interest return on your cash?

It's right around -- I think it's right around 2% or so.

That's after-tax, right?

Yes.

Shao Wang, Lotus Investment Management.

Hello. Two questions. First, I don't know if you gave a headcount number. I thought it was 756 at the end of the last quarter. I'm wondering if you have it for the end of the September quarter. Separately, on the comments at the beginning of the q-and-a you mentioned selling to -- regarding Checker-- selling to different customers. Have you a sense as to reverse (ph) (indiscernible) systems, have you a sense as to how long the sale cycle might be given the experience you have to date there for Checker?

The data I have is really spotty but I can tell you that because it's only $1000 people can make a decision -- people who want to make a decision make it very quickly. So we've sold demo units when we do presentations. People say, okay, I'll take the one there or they want to place -- we tell them to place an order and they say, no, no, no, I'll just buy it now. As a matter of fact, on a credit card -- on the company credit card.

So in some cases, the sale cycle was zero and in other cases I think what it's going to depend on -- you see, it's not a typical sale cycle where you're in competition with another company and you're vying to win the business. This is taking time with those customers who aren't yet prepared to switch from older technology to something that's brand new. So if you get it, if you understand it, you buy the product the day you see it. If you don't get it, you don't buy it. And I don't know how long it's going to take for us to convert those people.

If you don't get it and the customer stays with traditional sensors, if you will, do you go back and call on them again in three months time or is there a repeat effort?

I can't tell you that. I don't know. I don't know. My common sense would tell me if the salesman or the distributor would go in there, if the guy can't solve the problems with photo detectors. At some point, I'll tell some very interesting stories about low-tech companies trying to solve things with photo detectors and they couldn't, and they saw Checker and they bought it, and it's a company that oddly enough is a Mom and Pop shop that happens to have Wal-Mart as a customer. It's very strange, and they sell products to Wal-Mart -- I don't want to say what the product is -- and they are Wal-Mart's supplier for that product, and they had problems putting the product in plastic bags and filling up the plastic bags.

And there are millions of these, and it's Wal-Mart we're talking about here, okay? And they bought these machines, automation machines, to package their product, and the automation machines weren't functioning properly. They had photo detectors in the machines, and the distributor heard about this and we went in and retrofitted Checker, and that company immediately bought five that day. And this is a low-tech -- I can't tell you what the product is, but it's a candy kind of thing. So if you photo detectors don't solve the problem, then for sure we will get the order, but if people are uncomfortable, if they can kludge up a photo detector, I suppose, then we don't get the order yet.

But I will tell you that what's interesting, initially our estimates of the market size were based on complex applications where more than one photo detector was required -- actually, more than two photo detectors. If you needed more than two photo detectors, we base all of our market size based on that, those applications that needed two or more photo detectors. And that made quite a bit smaller market than the whole market for photo detectors. Well, strange but happily, most of our orders are to replace single photo detectors, because the single photo detectors are a little flaky. They weren't operating quite right. They were getting strange reflections perhaps, or the timing wasn't right. Photo detectors are very simple, very inexpensive, but they don't look very well in, it looks like, a lot of applications.

So we're happy to see that because what that means is the market for this product is greater than what we initially expected, and we thought it was about $500 million when we looked at the size of the current photo detector market, but it looks like the potential is greater than this. This product at the right price would become a photo detector killer. Now, it's too expensive to take over all the applications because if a $200 item can do it, then why would you spend $1000? But I believe that the market for Checker is extremely price elastic, and we are going to work on that.

And to answer your first question relative to the headcount, headcount at the end of the quarter was 740.

I'm assuming you're early on the budgeting process for '06. Do you have a sense or guess as to what headcount might be, say at the end of '06?

I have no idea at this point in time. You're very right that we are early in the process. I start my departmental reviews of the total company budget tomorrow morning at 9:00 A.M.

Okay, thank you.

But having said that, I would be surprised with any dramatic or even sizeable increases in headcount.

Yes, I don't see any sizeable. There will probably be some increases. I know we're going to be looking to add some incremental sales personnel in some of the international areas.

Now, we're talking without acquisitions, of course, because that changes the whole picture. But I would be surprised if we add more than, just the top of my head, 25 people, I'd be surprised. And I agree with Dick, that those additions would mainly be in sales and marketing. With DVT on board and with our existing engineering forces in Natick and in Portland and in England and in Japan, I would say that we are well-positioned. We have the engineering design team that's necessary. There might be a headcount here or there that we have to fill in, but primarily we're going to grow next year through increases in sales and marketing expenses.

Antonio Antezano with Bear Stearns.

Yes, a follow-up question. What are your plans in terms of commercial applications? You mentioned the activation of automatic doors, one of the applications before.

Yes. Unfortunately, that product isn't going to start shipping, I understand, until Q3 '06.

Then do you have like a target dividend payout ratio or anything like that? It seems like your average has been around 30% over the last few quarters.

Dick, I didn't quite understand it.

Whether you have a target dividend payout ratio.

Oh, oh. Ask it again. Very interesting question; ask it again.

I think the average for the last four quarters has been around 30%. Is that something that you expect to continue going forward?

That is a very interesting question, and I will hold my answer to that until some future date when you see a press release. Is that unusual, Dick?

Yes, it was --.

I mean it's unusual for me not to answer the question.

It was a little high over the last few quarters, mainly because the rate at $0.08 a quarter was established when our revenues were a little bit higher and when revenues came down as a result of the semiconductor capital equipment decline then our payout ratio turned out to be a little bit on the higher side because it did get in the high, high 30s and I think one quarter it was even close to 50% or whatever.

Now let me say that I am very pleased with the question because when we looked (indiscernible), what I couldn't understand was why people define dividends in cents. Why do they do that? I don't get it. And it must be because historically dividends are paid by nongrowth companies so the dividend always was a constant percent, not a constant number of cents. So your idea is the right one. And my idea are the same is that a dividend should be some percentage of the net income. If a company does well, the dividend goes up. If the company does poorly -- no, the dividend never changes. It's always the same percent. But the payout -- the amount of dollars that you pay depends on how successful the company is. That seems to me that's the right way to give a dividend.

Okay. Then in the surface inspection business, the gross margin for that business was below the other -- of your other two businesses.

For sure.

You mentioned that you were going to maybe put some plans to try to improve that profitability in that business. Is there any update on that?

That question was already asked and the update is that increasing gross margin in that business takes time. One has to -- there are only two ways, as I mentioned, to increase gross margin. You raise price or lower cost manufacturing cost and we would rather not raise price. That would be harder to win orders. We would rather lower manufacturing cost and that is what we're focusing on. That takes time because that means redesigning product.

(OPERATOR INSTRUCTIONS). Sarah Beyrich, Artemis.

I was wondering -- I guess on the DVT business starting with these new distributors or a new channel, will you still be pursuing other distributors or are you going to just kind of work with the people you have for now and see how that goes?

I'm not in charge of those decisions but I will tell you I know what the answer is going to be because people think the same way that we will constantly evaluate the quality and the results of our distributors and constantly update it.

But no significant (indiscernible) add thirty more.

No, I don't see that as happening. We did that significant thrust by buying DVT.

And then there is a small area and I forgot what you've called it but something like 3-D marking. Is that --?

Oh yes. It is called direct part marking, yes.

Anything further in that area that we may see --? Anything accelerating the growth there?

Oh yes, the growth is quite dramatic there and is going to continue to be dramatic. First of all, because this is an area -- the direct part mark reading is something that is mandated by a corporation for all its manufacturing. So it is not like each plant at Motorola or each plant at GE is going to decide whether to do it; they have to do it once the company says do it. So it is a high-level decision that is made corporatewide. And there are very few competitors in that business and we have (technical difficulty). I would say we became the major player in direct part mark reading. So Justin Testa runs that business and he is expecting rather solid growth and that is what we have seen for the last year to continue or maybe even accelerate.

Does that business get captured in factory automation?

Yes it does.

At this time, it appears we have no further questions. I'd like to turn it back to the management team for any closing remarks.

Thank you very much, Ian. It is almost time for me to go to bed. It is about 11:00 here. I have a busy day so I want to thank all of you for attending and certainly all of you who hung in for the end of the call and I hope to speak to you again about our results, which I expect will be very good for Q4. Thank you again and goodnight. And good morning.

Thank you. This does conclude this morning's conference call. Please disconnect your lines at this time and have a wonderful day.