Welcome to the TSMC’s third quarter 2007 results webcastconference call. Today’s event is chaired by Ms. Lora Ho, Chief FinancialOfficer and Vice President; and Dr. Rick Tsai, Chief Executive Officer andPresident. (Operator Instructions) Please be advised that those participants who do not yethave a copy of the press release may download it from TSMC’s website at www.TSMC.com.Please also download the summary slides in relation to today’s quarterly reviewpresentation. I would now like to turn the conference over to Dr.Elizabeth Sun, TSMC’s Head of Investor Relations for the cautionary statementbefore the main presentation by Ms. Ho and Dr. Tsai.

Good morning and good evening to all participants. This isElizabeth Sun, Head of Investor Relations for TSMC. Before we begin, I would like to state that management’scomments about TSMC’s current expectations made during this conference call areforward-looking statements subject to significant risks and uncertainties, andthat actual results may differ materially from those contained in theforward-looking statements. The information as to those factors that could cause actualresults to differ materially from TSMC’s forward-looking statements may befound in TSMC’s annual report on Form 20-F filed with the United StatesSecurities and Exchange Commission on April 20, 2007; TSMC’s registrationstatements on Form F-3 filed with the SEC on May 8, 2007; and such otherdocuments that TSMC may file with or submit to the SEC from time to time. Except as required by law, we undertake no obligation toupdate any forward-looking statements, whether as a result of new information,future events, or otherwise. Now I would like to turn the conference call over to Ms.Lora Ho, our Chief Financial Officer and Vice President.

Thank you, Elizabeth.Good morning and good evening to everyone. Welcome to our third quarter 2007earnings conference call. First, I will go over the highlights from our thirdquarter results. I will then give you the outlook for the fourth quarter.Please refer to the quarterly financial summary slides on our website. Alldollar figures are in NT dollars, unless otherwise stated. First, the highlights of the quarter. TSMC delivered astrong quarter in Q3, in which we set a record in terms of revenue and wafershipments. We expanded our gross margin and operating margin. In the meantime,we returned to $2.3 billion to our investors through cash dividends. Now, let’s turn to income statement. During the thirdquarter, we saw stronger than expected demand across major market segments. Asa result, we delivered revenue of NT$89 billion in the quarter, up 8% year overyear and up 19% quarter over quarter. Now, let’s take a closer look at our revenues. In terms ofrevenue by market segment, all three major segments delivered double-digitgrowth during the quarter, with the strongest growth coming from the computersegment, followed by consumer and communications. On a quarter-over-quarterbasis, revenue from computer, consumer and communications grew by 29%, 15% and12% respectively. In terms of revenue by technology, we continue to ramp our65-nanometer during the third quarter, as revenue from 65-nanometer more thandoubled from the previous quarter to account for 7% of wafer sales. We expectpercentage of revenue from 65-nanometer to further increase in the fourthquarter. Total revenue for Advanced Technologies accounted for 56% ofwafer sales, up 3 percentage points from the second quarter. Gross margin was 45.8% for the quarter, up 2.8 percentagepoints quarter over quarter, mainly due to increased wafer shipments and higherlevel of capacity utilization, offset in part by a decline in fuel price. Total operating expenses was higher than the previousquarter, largely due to higher legal fees and increased R&D spending on 55-nanometerand 65-nanometer related projects. Operating margin was 36.4% for the quarter. Income from non-operating items and long-term investmentstotaled NT$1.9 billion, down NT$1.4 billion from Q2, mostly due to accrual oflitigation losses related to the UniRAM case. Earnings per share were NT$1.15,down 7% year over year and up 19% quarter over quarter. Moving on to the balance sheet and cash flow statements. Wecontinue to have a strong balance sheet. We ended the quarter with NT$176billion in cash and short-term investments, down from NT$233 billion in thelast quarter, mostly due to an NT$82 billion payout of cash dividend andemployee profit sharing, offset in part by NT$26 billion free cash flowgenerated during the quarter. Now let’s turn to capacity and CapEx. Total installedcapacity for the third quarter was about 2.2 million 8-inch equivalent wafers.We slightly reduced our fourth quarter capacity, bringing our total expected2007 capacity to about 8.3 million 8-inch equivalent wafers, a 17% increaseyear over year. We spent $756 million in CapEx during this quarter. We nowexpect our full year 2007 CapEx to be around $2.6 billion, which include the$82 million purchase of 8-inch equipment from Atmel. With that, let me give you the outlook for the fourthquarter of 2007. Based on our current business and foreign exchange rateexpectations, we expect our consolidated revenue to come in between NT$92 billionto NT$94 billion. In terms of margins, we expect our fourth quarter grossmargin to be between 46% and 48%, and our operating profit margin to be between37% to 39%. That concludes my remarks today. Let me turn to our CEO, Dr.Tsai, for his remarks, followed by Q&A.

Thank you, Lora. Again, I would like to spend a few minuteson a few topics that many people have shown great interest in. The three topicsI am going to talk about include: Let me start with the business outlook. As Lora justmentioned, the third quarter was a good quarter for TSMC. We have performedbetter than we expected. I think the same can be said for the wholesemiconductor industry. All the PC shipments and handsets shipments I thinkexceeded expectations. We are now looking at the year 2007 in better shape compared tothree months earlier. Three months ago, we were forecasting a 3% to 4% growthfor the semiconductor industry in 2007. Now we’re looking at 2007 at 4%, andmaybe somewhat higher. The foundry sector, however, still will under perform thesemiconductor industry. We do not expect the foundry sector to grow much.Essentially, it will be a flat year for the foundry sector of the semiconductorindustry. Moving forward to 2008, what we are looking at right now isa better year from the semiconductor industry point of view. We expect rightnow for the 2008 to grow roughly between mid to high single-digits for thewhole semiconductor industry, while the foundry sector should outperform thesemiconductor industry. I think in the year 2007, inventory correction in the earlypart of the year -- first quarter very strongly and also to a large degree inthe second quarter -- impacted the foundry sector much more severely comparedto the semiconductor industry. Hence, quite a bit of lower growth for thefoundry compared to semiconductor. However, we are seeing the inventory effect to be quiteclear now. Third quarter inventory reports from most of the companies haveshown continuing improvement and this will continue into fourth quarter. Weexpect fourth quarter days of inventory to continue going down compared tothird quarter. So with the inventory effect gone, we believe the foundrysector will resume its previous pattern which is the outperformance over thesemiconductor industry. We certainly also expect TSMC to outperform thesemiconductor industry. Now let me talk about the CapEx. First let me comment on2007. Again, as Lora just said, our CapEx in 2007 will be about $2.6 billion,which is a lot of money. With this $2.6 billion, we will have acquired largechunk of the capacity that we will need in 2008. For example, the 90-nanometertechnology capacity we will be having, I believe, enough by the end of thefourth quarter this year and we probably will not need more 90-nanometer new capacitynext year. However, we will continue purchasing equipment for 65-nanometer and45-nanometer. The key point here is a large part of the capacity for 2008 willbe covered already by the $2.6 billion CapEx spend in 2007. In addition, we have worked very hard over the year on theproductivity improvement of all the tools that we have purchased, especiallythe 12-inch tools. We have improved, certainly, the productivity to the extentthat it also will help in lessening the need for the CapEx for year 2008. Essentially what we have done is to improve our CapExproductivity in 2007, hence we will be able to lessen the need for CapEx in2008. As a result, our current expectation for the CapEx in 2008 will be significantlylower than the CapEx in 2007. We believe this capital investment will allow usto support the business growth in 2008. Now let me go to the pricing part. We have always aimed toreflect the total TSMC value proposition. The value proposition includes thevalue of our long-term investment in technology development, manufacturingcapacity, design services, back end services, mask making services, et cetera,et cetera. However, our recent pricing trend was not consistent withour value proposition and the statement I just made one minute ago. We mustthen aim to keep the pricing trend in line with our value proposition. This isour plan for 2008 pricing. This is the end of my remarks. Thank you. We are happy toanswer any questions that you have.

Before you raise your questions, let me remind you to pleaselimit your questions to no more than two each time so that more people willhave their chance to raise their questions. Thank you for your cooperation.Operator, you can open the floor now.

Your first question comes from Shailesh Jaitly - NomuraSecurities. Shailesh Jaitly -Nomura Securities: Could you help clarify your comments in the prepared remarksabout the inventory situation? You had mentioned that inventories continue toimprove across the board. Do you see any pockets there? Do you still see excesses?Particularly if you can comment on the wireline communications segment, please.

Could you please repeat your questions again? Sorry, wedon’t have a very good sound system to completely hear your question. Shailesh Jaitly -Nomura Securities: I just wanted to get a bit more color on the inventorysituation across the board. I am wondering if there are any pockets of excesses,particularly if you can shed some more light on the wireline communicationsector with respect to inventories?

We do not see that. However, I want to caution that I’m notsure we have that fine granularity. I had said in my remarks earlier that weare seeing the inventory level in general go down at a very healthy level. Wecertainly expect to see that continue into fourth quarter. I might also add that from our output, or from our shipmentschedule and our customers’ business expectation, we do not see a concern ofinventory buildup. Shailesh Jaitly -Nomura Securities: So more or less across the board, it has improveddramatically and you don’t see any excesses anywhere?

I think we have seen a continuous improvement over the lasttwo to three quarters. So it’s a continuing process, I wouldn’t use theadjective dramatic as that would imply a sudden improvement. I think it is acontinuing improvement, and also we see that going forward, also. Shailesh Jaitly -Nomura Securities: I understand. Finally, if you could also comment on yourstrategy about the increase in stake at Vanguard, and whether you would be opento increase it beyond 50%?

Yes, we have increased our shareholding by 10 percentagepoints and reach about 37% of Vanguard’s shareholding. Our basic position withregard to Vanguard really has not changed. Vanguard has been and continues tobe our strategic partner in the mature technology business. We have worked veryclosely with Vanguard, both from technology covering all the way from 0.5micron to 0.18 micron. Vanguard also can cover certain business that we have notbeen able to cover, also. I think definitely a win-win situation. The increaseof the shareholding further solidifies the relationship. Of course, as of now,I cannot comment. I mean really, we do not have a firm plan as to what we willdo in the future as far as the shareholding is concerned.

Your next question comes from Timothy Arcuri - Citi.

Rick, can you be a little more specific about what you thinkthe ‘08 CapEx would be? When I look this year, you grew capacity 19% last yearwith similar CapEx. You grew capacity 24% the year before with similar CapEx.You grew capacity 17% this year with similar CapEx. It would seem that is not unreasonable that CapEx might comedown 50% or so next year. Is that a reasonable number?

Let me be straightforward. 50% is what you just said?

50%, yes.

I won’t say that’s unreasonable because I don’t want tocreate any confusion, but I will not comment further on any other numbers, allright? From anybody, not just you. But you have calculated a lot of numbers that we have also,but to be very frank, we continue to work on the numbers and we really haven’tmade the final decision yet but we’re fairly sure that the number will besignificantly lower.

It sounds like you spend a lot on lithography this year andalso last year. Is there some particular area of your capital spending whereyou think you’re getting a lot of efficiencies? Is it in litho, is that whereyou think you can scale back the most in ’08? Or is it something else?

We have made progress in general, but litho is definitelyone area that we have. We have put in a lot of effort during the last about 18months, for the simple reason that it is the most expensive part of theinvestment. But in general, we look at the more bottleneck type of machine. Becausethere is a lot of work to do, we try to prioritize them to ensure we spend ourresources most efficiently. But litho is definitely the number one priority.

Your next question comes from Pranab Kumar Sarmah - DaiwaSecurities. Pranab Kumar Sarmah -Daiwa Securities: Rick, you have mentioned about your CapEx productivity hasbeen improving. Do you have any particular target to add 1,000 more 12-inch fabcapacity in 2008, what type of CapEx you have to put in? Or if you can give ussome number like whether that CapEx productivity will go up by 5% to 10%? Anysort of target you have.

I think you used 5% to 10% and that’s probably not a badrange. We certainly target improvement in productivity. As I said earlier, wedo not try to improve productivity of every tool. What we are doing is toimprove productivity with certain tools so we can increase overall outputcapability from the same set of tools. I think 5% to 10% is, well, a fairlygood range to work with. Pranab Kumar Sarmah -Daiwa Securities: Because of your significant CapEx plan in 2008, do youforesee any market share loss in the foundry industry in 2008, or you will beable to maintain your desired market share?

We did not say that we will not have capacity. We believe wewill have enough capacity to serve the needs of the business requirements. Wecertainly will keep our market share. Just remember the revenue= price X quantity.

Your next question comes from Mehdi Hosseini - FBR.

There is obviously a lot of confusion over your capacityrequirements for next year, and it seems to me that we should expect asignificant price increase next year given the significant decline in CapEx. Isthat pricing premium driven by your expectation that you are able to continueto increase your market share or keep your lead at 65- and 45-nanometerprocesses?

I hope we have not created confusion that we will have lowcapacity next year. I think we tried to explain at the very beginning that withthe capital investment we have spent this year, together with the productivityimprovement of the critical tools that we believe with the significantly lowerCapEx that we will still have enough capacity to support our businessrequirements next year.

I’m just going back to your prepared remarks thatsemiconductor industry revenue could be up of 5% to 10% next year, andconsidering where your utilization rate is above 90%, the productivityimprovement must be significant. That’s where we are trying to understand: howsignificant is this productivity improvement?

Well, we certainly aim to do better than 90%.

I think there is one comment some analysts made thisafternoon in our conference saying that even if TSMC were not to spend an extrapenny on CapEx for next year, with whatever we have in the fourth quarter ofthis year, capacity, if you annualize that, it already represented a 9%year-on-year capacity increase even without a penny spent.

Just for clarification, did you say this morning or duringyour local conference call that CapEx could be down by as much as 50%?

No, we did not say that.

I just wanted to clarify. Where do you see 65-45 nanometertape-outs? Given the kind of tape-out you see at these two technology nodes, isit below where 90-nanometer tape-outs were a couple of years of ago when90-nanometer was first introduced?

I think 65-nanometer tape-out has been strong. I think our65-nanometer business ramp has been certainly according to our schedule andsomewhat better than that.  I forget thetape-out number, but the tape-out already in late October, we should have morethan 100 product tape-outs for 65. For 45-nanometer, I think it’s now a bit early to talk aboutthe product tape-out; we do have one. However, I think it’s important to saycompared to the same time of the 65-nanometer and 90-nanometer, we are engagingwith I would say significantly more customers at the 45-nanometer technology.

Sure. Is that one tape-out in the computing PC-related, orcommunication, wireless?

Wireless communication.

Your next question comes from Randy Abrams - Credit Suisse.

I wanted to follow up on the 45-nanometer, maybe talk aboutsome of the new applications you are seeing coming there. You are seeing moreactivity at 65 and 90, do you expect the ramp to be faster or a greater amountas we go to 45 then we saw at 65 and 90?

The application that we are seeing for the 45-nanometer nowis broader than 65-nanometer, with more customers. I would say now it’ssomewhat early to predict the ramp, but what I can say is if this situationcontinues, I would probably expect, yes, at least the same or better ramp. Basically, if you look at our 45-nanometer offering and theschedule of availability, it represents a very good value for the customers.

I wanted to ask, given your comment about inventory dayscoming down in fourth quarter, what is your initial view of revenue trends intothe first quarter ’08? Also, with the macro uncertainty, have you seen any notablechanges in the order book, whether it is cancellations or rush orders?

Of course we are not guiding first quarter now. We do notsee now any abnormal patterns other than the usual seasonal patterns. The firstquarter has some seasonal pattern now, I think it has continued for a fewyears.

Have you seen any real changes in the order trends in termsof just cancellations or rush orders in either direction, just given the macrouncertainty out there?

We’re not really seeing cancellations, no. We still see somerequests for the pull in of the wafers. But I would not want to mislead you tothink that we have a very strong situation; it is normal and probably a little better than normal.

Your next question comes from Daniel Heyler - Merrill Lynch. Daniel Heyler -Merrill Lynch: Good evening. I have a follow-up question, Rick, on 300-millimeter.What percentage of your current capacity is capable of 65-nanometer? It’s 7% ofrevenue, but I was wondering what percentage of your installed 300-millimeter,roughly, is 65-nanometer capable?

Dan, we are increasing the capacity of 65-nanometer so wehave seen the revenue continue to go up. Other than telling you what capacityallocates at 65, I can say that a big part of our capital expenditure for thisyear and going forward for next year will be needed to support 65-nanometer. Daniel Heyler -Merrill Lynch: Is it fair to say that a big part of your productivity gainwould be a lot of these fabs here that are running fairly full will bemigrating forward from 90 to 65 so you have a huge die advantage, a die shrinkcapability there? Is that where you are seeing the biggest gain in productivityor are there other factors that I am missing?

I think the productivity improvement is across the board,both for 8-inch and 12-inch. On 12-inch, it seems we are still adding capacity,so the productivity improvement has been more [phenomenal]. As an effect, gettingto reach to economic scale the productivity improvement is quite significant,so that also covers the 65-nanometer and also 90-nanometer and 0.13 micron aswell. Daniel Heyler -Merrill Lynch: Will that be a greater productivity gain, say from 90 to 65,relative to 130 to 90? If so, why?

You can say that again mostly came from the area that we areadding capacity, because it is a fill effect. So in a sense, you are right,since we are adding capacity with 65, it’s probably true. Daniel Heyler -Merrill Lynch: As you look out to ‘09, I guess much of your capacityadditions here in the second half of ‘07 are for your ‘08 growth. It seems asthough the industry, clearly the growth rates as Rick talked about, are lessvolatile. Do you have an ability or increased ability to start to think about‘09, or is that just crazy? Or do you think the industry will be in a 5 % to15% growth range, even out to ‘09?

I think the direct answer should be no, we don’t really havethat kind of ability, for sure. Even for next year is what we forecast based oncertain of our mechanisms, processes that we have. If we have to look at ‘09,we will have to use the macro trend line for the semiconductor industry trendline growth. Daniel Heyler -Merrill Lynch: Trend line, but then less likely to see massive inventoryswings that you have seen in the past? Clearly there are less players, right?There are less producers. I am wondering if you’ve thought through what thatmeans in terms of the volatility going forward and if that will be driving yourreturns further up?

Dan, can you repeat your questions? We don’t quiteunderstand your point.

Your next question comes from Mark FitzGerald - Banc ofAmerica Securities.

I was looking at the mix of IDM and fabless business andover the last year that has declined for the fables. I was curious if thattrend would continue going forward? Is there any correlation between the pricing problems and thattrend?

I think you probably see a change, I think, in the first quarterof this year or fourth quarter of last year, I forgot. But mainly that changecame mostly because of the acquisition of ATI by AMD, so there is a specificreason. Otherwise the ratio between IDM and the fabless has been fairlyconsistent over the year. Mark FitzGerald -Banc of America Securities: Is pricing to those two segments similar? Do you do betterin the fabless than you do in the IDM?

I think the pricing is very much customer-specific. Iwouldn’t really comment, that’s such a big group. Mark FitzGerald -Banc of America Securities: Isthere any evidence that your fabless companies have slowed their traditionaltechnology shifts here as the cost of designing chips has gone up? That theyare staying at a technology node longer?

I think this point have been raised by many people over afairly long period of time. I think the dynamic we are seeing is you probablycan say that they are going to 90-nanometer but it seems to us that as timewent on people learned and as I said just now, 45-nanometer engagement has beenstrong; has been stronger than actually 55-nanometer. My thinking is at the endof the day, economics rule.

Your next question comes from Steven Pelayo - HSBC.

Your lower CapEx next year is going to do wonders for yourfree cash flow. I’m curious if this is just, do you think, a one-time stepdown? I am not going to try to pin you down on an absolute dollar number, butover the last three years you have spent about 25% of revenues on CapEx. IfCapEx is down about 20% next year and you grow about 15%, that ratio is goingto go down to about 18%. Do you think that’s a new run rate for the next three-yearaverage on CapEx as a percentage of sales?

I want to answer your question very carefully. (1) This isnot a one-off phenomena. (2) However, we also do not have -- at least now -- aset ratio which we will use for the coming few years. We still look at CapEx based on the business outlook as wellas the strategic thinking. You also need to realize the annual CapEx has thisartificial annual thing. When we need to build, for instance, a shell for a newfab, that will definitely add a significant CapEx, which will not turn intoproduction for a longer period of time compared to CapEx for the incrementaltools, for instance, for next year. I cannot really promise you a set ratio, but this isdefinitely also not a one-off phenomena, either.

Maybe if you could comment, just in general, about yourpricing strategies here. How much of it is really making your customersunderstand the value proposition that you afford versus really the fact thatcapacity is just going to be growing at a much slower rate and your utilizationrate is going to be running at a higher level and so customers just need tounderstand that it is going to be a little bit tighter? Are you focusing on itfrom the value proposition side, or just the utilization rate is runninghigher?

It seems that if you raise the question with the customers-- I think the reality is both. I mean for any customer, they will look at manyfactors. Especially large customers, a long-term customer who has worked withus, they have many factors they look at when they decide to work with TSMC longterm, which includes the technological capability, long-term capacity, support,assurance of supply, ability to ramp, a good reasonable die cost, all of thosethings. Of course price is always a function of supply and demand, which, bythe way, TSMC is only a part of.

Your next question comes from William Dong - UBS.

I wanted to follow up on the split between 8-inch and12-inch. How would you characterize the difference in demand growth goingforward between 8-inch and 12-inch? Obviously, Advanced Technology is alwaysthe main demand driver but how would 8-inch fit into this? Does that alsoaccount for the limited growth going forward?

I would say Advanced Technology is certainly the growthdriver compared to a mature technology. However the mature technology, we dohave 5 million?

Yes, 5 million capacity.

5 million plus 8-inch capacity, which certainly represents avery significant amount of our business. We continue to try to upgrade ourbusiness in that area, too.

Going forward, would you grow your 8-inch capacity at thesame rate as 12-inch, or a little bit slower?

I would say slower.

In the terms of the economies of scale in the Shanghai fab,is that fab right now running at a lower margin than overall business, or do wehope that the economies of scale there improves so that it can reach the parentlevel?

You are right. I think right now the scale is not sufficientto render that as efficient and as profitable compared to our other 8-inch fabsin Taiwan. Weare further expanding the capacity there with, for instance, the tools we justannounced to purchase recently. I think that will certainly help that fab to beas competitive going forward.

Your next question comes from Matt Gable - MillenniumPartners. Matt Gable -Millennium Partners: Do you have any color at all from customers about thegeneral end demand trends in Q1?

Other than -- I think maybe I am repeating myself -- nothingother than a normal first quarter seasonal pattern.

Your next question comes from Bhavin Shah - JP Morgan.

What is embedded in the outlook for CapEx for next year? Canyou say how much is the CapEx for 200-millimeter? Or put it another way; howmuch additional 200-millimeter capacity do you think you might need to add, ifat all? Just a ballpark would be helpful.

I think there will be, at most, some incremental investmentfor 8-inch next year.

Depreciation has started to rise slightly at a faster ratein 2007, but perhaps that pattern will change again in 2008. Any guidance onwhat percentage of increase in depreciation we should assume for 2008?

Bhavin, since we have not finalized the 2008 CapEx, we justsay it will be significantly less.  Iwould expect the depreciation increase will be less than 10%, as we comparethis year versus last year. The degree of increase will be smaller.

Your next question comes from Keril Suiskin – CA AssetManagement.

Another question on CapEx; actually a clarification on theone you already answered. The question was what particular areas of equipmentyou would consider priority, or what areas you would consider cutting? Youmentioned litho, so I am not quite sure I understand whether you mentioned itas your number one priority for new spend or whether you are saying that youhave already spent enough to have capacity for another 18 months?

I think you probably have misunderstood. We were discussingproductivity improvement of certain tools. I was asked whether we have workedon the lithography tools. My answer was positive, mainly because lithographytools are the most expensive ones, and usually the bottleneck.

So if you were going to cut CapEx significantly in 2008,what areas would you consider strategically important in terms of equipment?

Let me first say, I do not know. I do not have a list ofequipment that we need to purchase. But on the other hand, we have a certaintarget for the 65-nanometer, 45-nanometer capacity, and we have a group oftools, and we will buy the balance of the tools. But I just don’t have the listmyself.

Could you possibly comment on the segment strength in Q4 andperhaps beyond that? So you were saying I think in Q4 the growth will be drivenby PCs. Could you may be expand a little bit, and comment on first half ‘08?

For fourth quarter, we think the computer-relatedapplications will grow the strongest, followed by communication-related applications,while consumer applications will decline, this follows their seasonal pattern.

What about first half ‘08?

We have not separated them in such fine detail. I think wewould expect next year for PC unit growth to stay roughly the same as thisyear. Handsets will continue to grow, but not as strong as this year.

Your next question comes from Mike McConnell - Pacific CrestSecurities. Mike McConnell -Pacific Crest Securities: Not to beat a dead horse here, but on the significant CapExprovision next year, is this more driven by the productivity gains you haveoutlined, or is this a function or a strategy to essentially stabilize pricing,considering the comments that pricing is more supply/demand driven?

Mike, just as I outlined in my opening statement, thecombination of the $2.6 billion spend this year and the productivityimprovement, we basically believe we should be able to invest less to get thecapacity we need for next year. That’s really what we have been saying. Pricing,we must work out the value to ensure we get value of our services. Supply/demand,as I said also, we are just a part of the supply, so we are not able to reallytalk about that. Mike McConnell -Pacific Crest Securities: if you look at your back end assembly and test, they haveobviously dropped their capacity additions considerably to try to stabilizetheir pricing and tighten up supply. Do you feel that, given the pricingpressure maybe from an industry standpoint, that we have seen in the foundryindustry that you and others may start to tighten up on capacity allocations totry to firm up pricing, drop depreciation, and then obviously get a higherreturn?

Mike, I really can only comment on what TSMC plans to do,which we have said many times. There’s absolutely no way that we can comment orknow anything about our competitors. Mike McConnell -Pacific Crest Securities: We have seen a slower cadence to which your customers aremigrating to 65-nanometer today then 90-nanometer, for various reasons. What’syour confidence though, looking at 45-nanometer, that that cadence willaccelerate and maybe we’ll start to see the adoptions reflect more of what we haveseen at 90-nanometer? What’s the difference between 45- and 65-nanometer if I’ma customer trying to decide whether I migrate to it faster or slower?

First, we have seen a substantially a stronger engagementfrom a higher number of customers and if this situation continues, then I wouldsay 45-nanometer will be a strong technology node. Why? I think 45-nanometer, in our case, we accelerate thetechnology schedule, actually, for our GS processes. The 45-nanometer GSprocess basically is a 40-nanometer design node. We offer an extra 15%, maybe20% area reduction. I would think that is very attractive to customers, andthey will be available at the same time as the 45-nanometer was supposed to beavailable. Ditto, I think, for the low power part. We also have learned over the years to, for instance, to getthe IP environment consistent and in sync with the technology available at the time.Everything we have been doing to meet with the customers, to enhance theirconfidence, I think they have all worked out.

If I look at 65-nanometer, the slower migration wasn’treally your fault. It’s just the fact that there is a higher rate of complexityof designs, more integration so customers are taking a longer time to tape-outbecause it takes them a longer time to come out with their design. Do you feel at 45-nanometer that even though you have thisgoing on, with these more complex designs from your customers, that you stillcould see maybe a migration to that node that’s going to be faster than what wesaw at 65-nanometer?

Mike, I’m not sure I totally agree with what you just said.At 65-nanometer yes, it is a more difficult and more complex design. On theother hand, I must say I have been impressed with some customers’ ability to designand ramp and also to shrink. I mean, not everyone, but one thing is clear tome: it’s doable; quite doable. If you do your homework early, more completely,if you work with TSMC sooner, and you are utilizing our capability, it works. That’swhy we are seeing also more and higher intensity in the collaboration level.

Your final question comes from Satya Kumar - Credit Suisse.

Could you provide some color on the pattern of spending fornext year? I was looking at the last couple of years for TSMC and CapEx seemsto be concentrated in Q2 and Q3. Is that a similar trend that we should expectfor ‘08?

We have not totally worked that out yet. I think for thisyear it is slightly back end loaded. I believe for 2008, since we have said itwill be significantly less, I tend to think that it will be slightly front endloaded in that case.

Would it be 60/40, is that a good estimate?

I don’t have a number right now. We are still working on thenumber.

Will it be higher than the second half of ‘07, Lora, or are youlooking at maintaining the second half?

I don’t have that at this moment. Maybe we can tell you morenext quarter.

Ms. Ho, there are no more questions at this time.

Thank you everybody for your participation. We are lookingforward to talking to you next quarter.