輝達 (NVDA) 2008 Q3 法說會逐字稿

Good afternoon.

Thank you for holding.

I would now like to turn the call to Mr.

Michael Hara, Vice President, Investor Relations.

Thank you, sir.

You may begin.

Good afternoon and welcome to NVIDIA's conference call for the third fiscal quarter ended October 28, 2007.

On the call today for NVIDIA are Jen-Hsun Huang, NVIDIA's President and Chief Executive Officer; and Marv Burkett, NVIDIA's Chief Financial Officer.

Before we begin today's call, I would like to take care of some general administrative items.

Your lines have been placed on a listen-only mode until the question-and-answer segment of today's call.

During this call, we will discuss some non-GAAP financial measures about net income, diluted net income per share and gross margin, when talking about our results.

You can find a reconciliation of these non-GAAP financial measures to the GAAP financial measures in our financial release, which is posted on the investor relations page of our website at www.nvidia.com.

Unless otherwise noted, all references to research numbers throughout the call come from Mercury Research.

This call is being recorded.

If you have any objections, you may disconnect at this time.

Please be aware that if you decide to ask a question, it will be included in both our live transmission as well as any future use of the recording.

Also, shareholders can listen to a live webcast of today's call via the investor relations page of our website at www.nvidia.com.

The webcast will be available for replay until the Company's conference call to discuss its financial results for its fourth quarter fiscal 2008.

During the course of this conference call, we may make forward-looking statements based on current expectations.

Forward-looking statements, including statements as to the importance of and uses for GPUs; our outlook, the impact, features, performance and availability of our products and technologies; our strategies; our growth and growth drivers pertain to future events and are subject to a number of significant risks and uncertainties.

The Company's actual results may differ materially from results discussed in any forward-looking statements.

For a complete discussion of factors that could affect the Company's future financial results and business, please refer to the Company's Form 10-Q for the period ended July 29, 2007 and the reports on Form 8-K filed with the Securities and Exchange Commission.

All forward-looking statements are made as of the date hereof, based on information available to us today, and except as required by law, the Company assumes no obligation to update any such statements.

The content of the webcast contains time-sensitive information that is accurate only as of November 8, 2007.

Consistent with the requirements under Regulation FD, we will be providing public guidance directly in the conference call and will be unable to provide significantly more information in off-line conversations or during the quarter.

Therefore, questions around our financial expectations should be asked during this call.

At the end of our remarks, there will be time for your questions.

In order to allow more people to ask questions, please limit yourself to one question.

After our response, we will allow one follow-up question.

I will now hand the call over to Jen-Hsun.

Thanks, Mike.

Good afternoon and thank you for joining us.

Today, we're pleased to report record revenue, record net income and record gross margin.

This is also our first billion-dollar quarter.

For our third quarter, revenue grew approximately 19% sequentially from Q2 and 36% from last year to a record $1.12 billion.

GAAP net income increased more than 121% year over year, and GAAP gross margin improved by 550 basis points from a year ago to a record 46.2%.

Let me review some of our recent achievements.

As the first GPU Company to reach the billion-dollar quarter level, NVIDIA continues to be one of the fastest-growing semiconductor companies in the world.

Our GeForce desktop and notebook GPU product lines each achieved record revenue.

The desktop GPU product line grew 33% year over year, and the notebook GPU product line grew 120% from last year.

Our stand-alone notebook GPU share grew to 72%, up from 52% a year ago.

In September, we shipped our first-ever single-chip motherboard GPU for Intel processor-based desktop PCs, the NVIDIA GeForce 7000 GPU family.

The GeForce 7000 family delivers the performance of an entry-level discrete GPU and leading graphics compatibilities when compared against traditional integrated graphics solutions.

We estimate the Intel segment represents approximately 40 million units of new addressable market opportunity next year and expect it to be a strong growth driver for our GeForce motherboard GPU business.

It October, we launched a new GeForce 8800 GT GPU, the first in our family of 65-nanometer desktop GPUs.

The 8800 GT combines best of gaming performance and best in HD playback at the $199 price point.

Reviewers are calling the 8800 GT the next GeForce4 Ti 4200.

The Ti 4200 was called the perfect GPU of its time, because it was simultaneously modern, high-performance and well-priced.

The Ti 4200 was one of the most successful products in our history.

The widely recognized tech website AnandTech wrote, "It's really not often that we have the pleasure to review such a product so impressively positioned.

The 8800 GT is a terrific part, and it's hitting the street at a terrific price.

The performance advantage and price utterly destroys our perception of the GPU landscape."

Our Quadro Professional Solutions business achieved record revenue and grew 37% year over year.

During the quarter, we launched seven new Quadro solutions and, most notably, the Quadro FX 370 and 570, which deliver ISV-certified hardware at sub-$200 price points.

The computer-aided design industry is undergoing a massive transition from 2D illustration to worldwide collaborative [3D] design and could represent tens of millions of new CAD seats over the coming years.

We believe the FX 370 and 570 are ideal solutions for the entry-level designer, and will continue to accelerate our growth in the Professional Solutions segment.

Finally, we begin shipments of the Tesla C870 GPU computing processor and D870 deskside supercomputer products this quarter.

The adoption of Tesla and the CUDA C-language programming environment and tool suite continues to accelerate.

We believe we now have more than 3,600 developers in 14 industry categories working with CUDA on exciting applications ranging from automotive vision systems, biological simulations, computational fluid dynamics, seismic processing and imaging, speech synthesis, computational finance, quantum molecular dynamics, protein folding to computer vision.

Let me turn the call over to Marv to discuss our financial results in more detail.

I will return in a moment to discuss our businesses and growth drivers for Q4 and beyond.

Thanks, Jen-Hsun.

Today, we will be looking at both GAAP and non-GAAP results for our Q3 and nine months ended October 28, 2007.

The only difference between GAAP and non-GAAP is stock-based compensation and its tax effect.

As Jen-Hsun stated earlier, revenue for the third quarter was $1.12 billion, which is up 19% quarter to quarter and up 36% year to year.

Obviously, it was a very strong quarter, and our concerns at the beginning of the quarter as to the sustainability of the strength we saw in Q2 never manifested.

Also at the beginning of the quarter, there were concerns relative to production limitations.

Through the efforts of our partners and our operations staff, we were able to overcome most of those limitations.

In a few cases, our revenue was limited by production, but relative to our concerns at the beginning of the quarter, these were minor.

In our GPU business, the growth was led by notebook, which grew 41% quarter to quarter and 120% year to year.

The desktop business grew 14% over the second quarter and was up 33% over the prior year.

In the quarter, GeForce 8X products accounted for more than 80% of our desktop and notebook business.

Our MCP business grew 23% quarter to quarter, as we begin to see traction in our new Intel platform products.

PSB, our workstation business, grew 20% over the second quarter and 37% over the prior year.

Even our consumer business grew 6% over Q2.

All in all, there was strong growth in all segments.

On gross margins, we reported GAAP gross margin of 46.2%, which is up 90 basis points from Q2.

Non-GAAP gross margin was 46.4%.

We had anticipated strong gross margin based on the growth in GeForce 8X products, the increase in professional solutions and the increase in PlayStation 3 royalties.

All of these contributed to the increase in gross margin.

We are again pleased with our progress.

Operating expenses were $268 million GAAP and $238 million non-GAAP.

This was slightly higher than our guidance.

A major factor in this increase was the improved performance and outlook for the whole fiscal year and how that affected our variable compensation programs.

For the quarter, we had to accrue an additional $8 million as a catch-up based on the increased financial performance and expectations.

This is a one-time catch-up.

We ended the quarter with 4,609 employees, which is up 236 from the second quarter, as we continued to aggressively hire, particularly in R&D.

155 of the 236 new employees were in R&D.

Depreciation for the quarter was $33 million and capital expenditures were $70.4 million.

The increase in capital expenditures can be attributed to two factors.

We bought additional testers of $22 million.

These testers are consigned to our contractors in Asia for the increased production.

We also purchased some land for $27 million in anticipation of building additional space.

Interest and other income was $19 million in the quarter, and the tax rate was adjusted down to 13% for the year to date.

This resulted in a tax rate for the third quarter of 11.7%.

The slight downward adjustment results from more international sales than anticipated, as well as higher R&D tax credits.

On the balance sheet, cash and marketable securities ended the quarter at $1.85 billion, which is an increase of $281 million from the second quarter, despite repurchasing $125 million of stock during the quarter.

Operating cash flow hit $400 million in the quarter for the first time.

Accounts receivable were $552 million, an increase of only $44 million from the second quarter, as the even profile of shipments during the quarter helped collections.

DSO for the quarter was 45 days, down from 50 days in Q2.

Inventory grew slightly to $306 million, in spite of the strong sales and production limitations.

Most of the increase can be attributed to new products that were not introduced until after the end of the quarter.

DSI for the quarter was 46 days.

On the outlook, we've had two consecutive excellent quarters, where revenue has exceeded our expectations.

Even with this, we currently see strong demand for our products, and as a result we expect revenue to grow in our fourth quarter.

Revenue growth in the 5% to 7% range would be within our expectations.

We expect our growth to be led by the continued strength in GPUs and new products in both GPUs and MCP.

For gross margin, we have made excellent progress and will work very hard to maintain our gross margin.

New products will help, but we are currently ahead of our plan in our progress.

Operating expenses will increase in the fourth quarter, despite the catch-up that occurred in the third quarter.

This increase is the result of additional hiring and increased R&D expenses associated with new products.

A total operating expense increase in the 3% to 5% is reasonable.

We would expect the tax rate to stay close to our 13%.

But if there is any adjustment, it would have a downward bias.

In summary, we expect an excellent Q4.

I will now turn it back over to Jen-Hsun.

Thanks, Marv.

We are entering the era of visual computing.

Some of the most exciting applications require a GPU for the best experience.

Industry expert PwC estimates the worldwide videogame market to be $30 billion in size and expects the market to grow to $50 billion by 2010.

According to comScore, 217 million video gamers immerse themselves in online virtual worlds, and expect that number to grow 17% annually.

To date, there have been over 200 million unique downloads of Google Earth.

SketchUp makes it easy for millions of users around the world to convert Google Earth into a full 3D world.

Microsoft has shipped 85 million units of Vista, and Apple's news Leopard looks amazing and is a step forward in UI design.

Both demand GPU to deliver the best experience.

Our computing experience is increasingly enhanced by the GPU.

As the largest GPU company and the world's leading visual computing company, we're very well-positioned to continue to lead this revolution.

Near-term, we see multiple growth drivers.

First, GeForce discrete GPU -- GeForce discrete GPU grew to 21 million units in Q3, up 30% from a year ago.

The combination of our technology leadership, great content support and the ever-increasing importance of the GPU in our computing experience are contributing to drive our growth.

We estimate the installed base of GeForce to be approximately 200 million units.

That makes GeForce the largest game console community and a target-rich environment for game and application developers.

We believe the PC game market is continuing to grow.

Our overall enthusiast GPU segment grew 35% year over year.

The GPU enthusiast GPU market grew 170%.

Our strategy is to make GeForce one of the most important processors of the visual computing era through technology leadership, increasing programmability, great content experience and promoting our great brands.

Second, GeForce motherboard GPU -- GeForce motherboard GPUs grew to 10 million units per quarter, up 50% from a year ago.

The GeForce 7000 is off to a great start.

We are now the only chipset supplier to support both Intel and AMD processor platforms, and the only branded integrated GPU supplier for the Intel processor platform.

The integrated graphics opportunity represents approximately 60% of the world's PC market.

Our strategy is to bring the benefits of GeForce to the most sensitive price segments while creating exciting platform architectures like SLI, Hybrid SLI and ESA.

ESA, which was recently announced, stands for Enthusiast System Architecture, is a standard for system information protocol that links the system's various critical components such as fan, power supply, smart chassis, GPU cards and motherboards.

It enables a unified architecture for applications and users to control and optimize the performance of their system.

SLI, Hybrid SLI and ESA are examples of how NVIDIA creates architectures that advance the capabilities of the PC.

Third, global proliferation of industrial design is driving Quadro growth.

NVIDIA's Quadro brand is the de facto standard for workstation visualization.

Quadro is not just a GPU.

Quadro is a professional class of GPUs that incorporate features demanded by creative professionals, as well as software that completes a solution need for each vertical market, whether auto, medical imaging, film or financial markets.

We are experiencing multiple simultaneous dynamics in the professional market.

Traditional 2D design and illustration is transitioning to collaborative 3D design and product lifecycle management.

Globalization is forcing multi-site design teams around the world to adopt common platforms and is sponsoring the growth of local industries in China and in India.

All these dynamics are expanding the workstation TAM.

Fourth, GPU computing with CUDA and Tesla -- CUDA is a data-parallel programming environment built on C.

CUDA runs on all the millions of G80-based and beyond GPUs, whether Quadro, GeForce or Tesla.

Tesla is the brand of GPU software system solutions that's targeted at the high-performance computing market.

The initial adoption of Tesla will be inclusion into supercomputing clusters.

Next year, we should see deskside personal supercomputing workstations powered by Tesla for data parallel supercomputing and Quadro for interactive visualization.

With CUDA on GeForce, we will quickly reach a 100 million unit per year run rate for CUDA-powered GPUs for the consumer market.

For consumers, CUDA on GeForce will enhance applications like video and image processing and physics processing for games.

CUDA is one of the most revolutionary and disruptive ideas from our company.

CUDA has ignited interest on GPU computing all over the world.

There is growing understanding that the data parallel architecture of a GPU is radically different than that of a CPU, and for a large class of compute-intensive data-parallel applications, the GPU can increase performance from 10 to several hundred times.

We expect our first application processor to sample at the beginning of next year.

We believe another personal computer revolution is about to happen.

Mobile devices like phones, music players and portable navigation devices will all increasingly become multifunction, multitasking personal computers.

If so, the architecture of these devices will increasingly become consumer PC-like, and be capable of delivering all the entertainment and Web experiences we enjoy on our PC today, but in a form factor that fits nicely in your hands.

Our mobile strategy is to drive the second personal computer revolution by creating a from-the-ground-up application processor and system on a chip that enables this experience.

[AP15], as it is called internally, is the culmination of several hundred man-years of R&D.

The mobile application processor is an area where we can add a tremendous amount of value and represents a several hundred million units a year growth opportunity.

We're happy to take your questions now.

(OPERATOR INSTRUCTIONS).

Hans Mosesmann, Raymond James.

Congratulations, guys.

The GeForce 7000 motherboard GPU -- what kind of average selling price are we talking about here, and are they consistent with the average selling prices of the AMD platform version?

The ASPs on the MCP73 or the GeForce 7000 is slightly higher on the Intel side.

But it should be targeted -- we're targeting that product to be in the mainstream price-sensitive segment, so that we can bring GeForce to as many people as possible.

That's a rather large segment, and we think it represents about 40 million units of incremental opportunity for us next year.

Is that a $50 ASP, $100 ASP?

I just want to get a general sense of the magnitude.

No, it's in the teens; it's designed to be on the motherboard.

It's intended to be in the teens, and we want it to be as affordable as possible, so that we can address the largest possible market segments.

Thanks a lot and congratulations.

Heidi Poon, Thomas Weisel.

Can you describe what you expect further for notebooks?

I know the general trend of desktop to notebook transition is still there.

But most of your wins have been in discrete.

As AMD and Intel come up with their integrated graphics platform also supporting DX10, where do you think you can take your market-share gain through next year and after that?

Well, it's really hard to say what our market share will be in the future, but we've competing against integrated graphics for quite a long time.

Our ability to succeed will depend on continuing to do the types of things that we do well, which is come up with great ideas like SLI and Hybrid SLI and bring value to the marketplace like we do with GeForce and Quadro.

So we're going to continue to do those things.

If we do our jobs well, we should continue to grow.

If we don't, then obviously the competition will do better.

But do you think that you might have a disadvantage because of the power consumption factor?

How is their integrated graphics lower power consumption than ours?

In fact, I think that you know, probably, that NVIDIA's G8x architecture is the most efficient performance per watt GPU in the marketplace today.

That explains why we're able to get our enthusiast-level GPUs into a single slot with a single-slot fan, why our fan noise is substantially lower and obviously our measured power to be substantially lower.

That's something you can do with architecture.

That's something you can do with innovative circuit design techniques and so on and so forth and system technologies to manage it at the system level.

So there's all kinds of competitive capabilities for us to innovate in power, and we believe we're absolutely world-class in that area.

Shawn Webster, JPMorgan.

Can you give an update -- a lot of folks, especially investors in the PC chip-related companies, are concerned on linearity, risk of double ordering and things being too good for too long.

Can you guys share with us some of your thoughts?

I think, Marv, you had some comments on linearity.

Maybe you could talk about order linearity as the quarter progressed, how your backlog looks and any risk that you can tell in terms of double ordering?

Our Q3 was one of the most linear that we've ever had.

The strength that we saw in orders in Q2 continued straight through Q3.

So if you wanted to look at a backlog, we're not a backlog-focused type company, but there was no change.

The orders and the shipments were ratable all the way through.

Obviously, you're always concerned about is there any double ordering going on.

But we don't see any evidence of that.

The way I would think through that problem is, first of all, we don't know what we don't know.

Obviously, nobody does.

The marketplace is rather robust, and I think it's rather robust for us for a variety of reasons.

I think, first of all, is new products always helps, and the 8800 GT is arguably the best product that NVIDIA has built in five years.

That's what everybody is telling us; that's how we feel about it.

We have very, very big hopes for it.

We have MCP73, which is our first product of that category and bringing GeForce to the Intel motherboard is pretty exciting stuff, and the early adoption has been wonderful.

Both of us products we just can't make enough of.

There's good reasons for their growth.

I think overall, the GPU dynamic is pretty clear now.

PC games, online games is one of the fastest-growing application sectors.

Vista is doing well.

Leopard is amazing.

There are more and more applications that, in combination with a GPU, is just a better experience.

Then we have new innovations like Tesla and CUDA.

There's a very logical reason why there's industrial growth in design.

When we globalize to China and India and other Eastern European countries and we collaborate across the network and we design 24 hours a day, it makes perfect sense that you would need compatible workstations all over the world.

Simultaneously, some of these economies are no longer just manufacturing economies, but they are moving into design economies.

So, as you know, China is moving very, very quickly up the food chain and more and more unique designs are coming out of China.

Well, they need workstations for that.

So I think, if you combine whether it's new products that we're introducing in the market that we serve today, the importance of the GPU, new industry growth from workstations or a new disruptive technology that we've created with Tesla and CUDA, those are all important reasons why we're growing.

You guys mentioned that you had a couple areas that were still tight, from a capacity perspective.

Could you share with us what areas those are?

Jen-Hsun just mentioned a couple of them, the CP73 and some of our GeForce 8 products were capacity limited or production limited in the quarter.

(OPERATOR INSTRUCTIONS).

Simona Jankowski, Goldman Sachs.

I just wanted to get a better sense on your Q4 guidance.

How much of that is based on your expectations for further share gains versus the end market growth or other factors such as potentially ASP increases?

We don't count on share growth.

That happens after you are done counting it.

So unless it's a design win or whatever it is that we know, we don't typically count on share growth.

So we look at the strategic position of our products, we look at the new products that we're introducing, and therefore we have got some confidence on its pricing power.

We count on new markets that we're entering, opening up.

So we count on a variety of things.

Of course, OEM design wins are OEM design wins.

But otherwise, we try not to be too bullish on market-share wins.

We also don't anticipate any changes in ASPs in the various market segments.

That's not [in the] growth.

So it sounds like a lot of it is just the end market continues to be very robust.

Yes.

And we're entering new markets.

We're creating new products for existing markets as well.

When we start to ship Tesla, that's a product that's the first of its kind.

It adds so much value, and the people that buy it initially are people who are desperate to have more computing resources.

Products like the 8800 GT -- it's in the existing market that we serve, but it's very disruptive because of its price performance is just so terrific, and it really promotes very, very quick adoption.

Oftentimes, it even encourages pretty broad-based upgrades.

You know that there's a whole bunch of games coming out this Christmas with DX10, and the games look really terrific.

So there's just a lot of reasons why the GPU is doing well.

I appreciate that color.

Just one other question that's somewhat related.

Gentlemen, you mentioned that you think the motherboard GPU is an incremental 40 million unit opportunity.

These products seem to be competing very effectively against the lower end of the discrete GPU market and seem to have a significantly lower price point.

So do you view this as an incremental opportunity, or do you view this to some extent as a substitution of some of the existing stack of the product?

Well, the GeForce 7000 really isn't high-performance enough to be a replacement for even our entry-level GPU.

So that's not -- and you know our discrete GPU -- as, I think, one of the data points that Marv just mentioned, our discrete GPU will be vastly converted to DX10 very, very shortly.

So this is really targeting the segment that's underneath it, the 60% of the marketplace that we don't really serve today.

Gurinder Kalra, Bear Stearns.

My first question is on the GeForce 8800 GT.

It's a great product, but do you think there's any chance it might be cannibalizing some of the very high-priced products?

Given that it's priced from $200, $249 but on 65-nanometer, how does it affect your gross margins going forward?

Well, we are always trying to be thoughtful about how our products can cannibalize our own products.

But I think we understand this market pretty well, and if it's possible to cannibalize our products, well, I'd rather do it myself than wait around for somebody else to come and do it.

So I'm not too worried about that.

That's really just the nature of our business, and we have to stay aggressive.

We want to stay aggressive, and we want to bring as much as we can to the market.

So I'm not inclined, really, to slow down because it somehow protects my own business.

My own business will have to figure out a way to deal with my own business.

So that's our strategy there, nice and simple.

But I think our high-end products will continue to be best of its kind, best of its class, and I think the market will be fine there.

As far as margins go, the margins on the 8800 GT are excellent, well above corporate margins.

Just a follow-up question on the general-purpose GPU/Tesla.

Can you discuss some of the traction or the design wins you're seeing in the financial services industry and the headway you've made there?

There are a lot of development going on using GPU computing for computational finance.

As you guys know, your industry has some of the brightest mathematicians in the world, and the amount of computation you guys need is completely unbounded.

You just don't seem to be able to get a large enough server or enough of them.

We've demonstrated that GPU can be used to accelerate computational finance by 10 times, 20 times and, in several applications, 50 to 100 times.

That's a very, very large step up in performance improvement, so there is quite a bit of development in that area.

It's happening all over the world.

It's happening in London.

It's happening in New York.

This is an area that I think is going to be pretty interesting for GPU computing.

Michael McConnell, Pacific Crest Securities.

On the G92 or the 8800 GT, when are you expecting crossover on that relative to the 8800 GTS?

Crossover?

Let me --

In terms of when it will be the majority of shipments in the midrange category or performance mainstream.

Let me explain it this way.

I think this will do it.

We will ship more 8800 GTs this quarter than we shipped last quarter of 8800 GTS.

So this will be (multiple speakers).

Would you say this is your probably most aggressive ramp in history, maybe, of a product?

This is one of our most aggressive ramps, and the reason for that is because it's such a great product.

We have a really rich backlog of orders for 8800 GT.

The market now knows what the competition has, and our customers know what the competition has, and they are voting with their orders.

We are going to be very, very aggressive with 8800 GT.

It's just such a great product.

Five years ago, we had the 4200 Ti, Ti 4200.

The market called it the perfect GPU.

The reason for that was because it was a brand-new architecture, it was extremely high-performance and it was very affordable; it was $199.

It lit that segment on fire, and we couldn't build enough of that 4200 for a very long time.

So we think we've got another 4200 on our hands, and certainly the market has effectively called it that.

It's tough to predict market share.

You guys have obviously dominated this segment in the high end.

Can you talk about just the basis for having such an aggressive ramp?

Do you really feel like you can take more market share, or is this just you are ramping it in accordance to the demand out there, which is quite strong?

There's always competition in our space.

The GPU industry is one of the most vibrant industries, and the reason for that has to do with the fact that Moore's Law is still GPU's friend, or GPU is still Moore's Law's friend.

This is one of the few technology segments left where more transistors and new algorithms and new technology could enable better experiences.

Almost all of the other semiconductor sectors are seeing smaller die sizes every year and lower ASPs every year, but not the case with GPU.

So I expect this to be an area of quite a bit of vibrance for a decade to come.

In terms of how quickly we ramp, that's just the way we do things.

We need to bring out new architectures and we need to bring it out as broadly as we can.

Because as soon as we launch one unit of 8800 GT, everybody in the world is going to want it.

So you might as well assume that you are building for everybody in the world, and so that's the approach that we've taken.

We want to stay aggressive as a company; that's just our nature.

Then a final question, just on the GeForce 7 series chipsets for Intel's platforms.

Can you talk about the early traction you're seeing, both at OEMs and then maybe separately in the channel and just how that is progressing and the outlook, if you have one, for next year?

We're expecting to be very successful with OEMs as well as the channel, and so it's designed to -- NVIDIA's brand and the quality of our technology and the rich support that we offer in drivers -- as the industry gets more complex, with DX10's and OpenGL's and high-definition videos and all of these different applications, our software capability becomes an increasing advantage.

Just because it's integrated on the motherboard doesn't mean the software isn't complex, and that's an area that we are extremely good that.

So my sense is that we will be very successful in both OEMs and the channel, and you'll see it soon.

Marv, have you guys thrown out just a target for next year within that business, or just quantified what the opportunity could be for next year?

I don't know if you have or not.

No, we have not.

You don't care to, or --?

I believe it's a significant growth driver next year.

But no, we haven't quantified it.

Tayyib Shah, Longbow Research.

Congratulations on the quarter.

The notebook market has been doing very well lately, but maybe partly at the expense of desktop segment, where you have higher exposure.

Can you help us understand how desktop replacement by notebooks is likely to affect your business going forward, now that you already have 70% plus share in notebooks?

The notebook market is certainly doing very well, and I'd expect that one of the areas where the notebook market is just starting to see growth, but it will see a lot more growth in the coming years, is the gaming notebook segment and the mobile workstation segment.

These two segments are growing very quickly, and I expect that we'll be able to add a lot of value to those two platforms in the future.

We're not too hung up about where people buy our GPUs.

We are delighted by the proliferation of platforms.

Whereas people typically tend to have one or two desktop PCs in their house, they may have four and, in my case, eight laptops at home.

So it's a little bit like a cellphone, I guess.

You just buy a whole bunch of them over time.

So I think that the notebook market is not completely a replacement market, frankly.

I think that a lot of it is an additive market, and certainly it has exceeded most people's expectations.

Another question along the same lines.

As your footprint within the notebook market has grown, have you been finding this market more competitive than your discreet GPU business, given that in this business Intel can bundle their chipsets with their CPUs, if they want to?

Intel is definitely a very competitive company, and we've been competing against bundling for, I guess, quite some time, both, I guess, whether it's price bundling or other bundling.

The bottom line is we just have to add value.

We just have to add value.

NVIDIA is not a commodity player, and we serve the marketplaces where visual computing is really important.

If 3D matters to you, and if your work matters to you and you do 3D work, whether it's in design or film or medical imaging, then we could add a lot of value to you.

If your business really depends on having a great GPU and visual computing capability, I think NVIDIA is really your partner.

(OPERATOR INSTRUCTIONS).

Glen Yeung, Citigroup.

This is Peter Karazeris for Glen Yeung.

A question just on ASPs.

I was wondering if, on a mix basis, your ASPs were able to improve in desktops and notebooks from second to third quarter.

No, they were relatively flat.

Despite mix.

Then I noticed -- so margins were up, and you talked about, I think, royalties from Sony as being one of the drivers.

But consumer revenue was up about 6%.

I was wondering if you could give us any color around consumer revenue, and then also if you could just help us understand the total impact that royalties had on gross margin improvement.

Well, any increase in royalties is 100% gross margin.

So I would say that our royalty income from PS3 increased significantly from Q2 to Q3; it more than doubled.

So that had an impact on our gross margins.

Was it the driving factor?

Probably not, but it was one of the factors that improved gross margins.

The fact that that increase would probably account for most of the increase in the consumer business is probably true.

Uche Orji, UBS Investment Bank.

Could I just get a sense of where the Professional Services business will go next quarter?

I see your guidance and you are [holding] (technical difficulty).

But that division is pretty critical for driving overall gross margins.

So if you can give us some color as to what we should expect within that division, that would be very helpful.

The second question is about your handset strategy.

I understand your push on the apps processor direction, but do you see the lack a baseband solution as something that could become a problem for the execution of that strategy?

So if you can give me some color, that would be helpful.

I guess, on the first one, we really haven't, and we don't give too much granular guidance at the business unit level.

All I can tell you is that the Professional Solutions Group is doing well, and they're strategically very well-positioned.

They add a lot of value to the marketplace.

Our business, what some people call the workstations business, what we call the Professional Solutions business, is really a combination of GPU technology and a lot of software system component technology and expertise.

We have one of the -- if not, in my expectation, to be the largest developer technology organization in the world.

We help inspire and we help solve some of the most challenging visual computing problems in the world through that organization.

So I expect it to do well.

We haven't really given much guidance.

With respect our application processor business, we're going to focus on segments of the handheld marketplace where computing is particularly important.

If you take a look at all the handhelds in the world, really with the exception of the iPhone, every other phone in the world wants to be a computer but has a long way to go.

This is where we could add a lot of value.

NVIDIA is fundamentally a computer technology company.

We are one of the world's largest computer technology companies, and there are very few consumer-oriented computing platforms where we can't, from the ground up, design an entire computer for you.

So this is an area that we could add a lot of value.

We're seeing a lot of interest in the work that we're doing.

When I said several years ago that this is going to be the second personal computer revolution, and that the phone is going to become a computer first and a phone second, at the time it sounded kind of odd.

But I think it makes a lot of sense now.

(OPERATOR INSTRUCTIONS).

Daniel Ernst, Hudson Square Research.

First, on the mainstream growth or maybe also the low end of the discrete, obviously, in the domestic markets so much enthusiasm around digital video and high-resolution photography with things YouTube and digital SLRs and the uptake of premium Vista -- I assume that has been a lot of the growth there.

But can you talk about that same market overseas, where so much more of the overall PC market growth is, especially in developing countries like India and China, where I assume the market is more price-sensitive?

Then secondly, on the mobile phone business, you gave us a little bit of mapping for the app processor.

But can you give us a little bit of color about the current mobile phone business and how that is tracking relative to expectations?

I would say that the developing markets are being driven right now by online games, which is really a craze in China and India, and the online game market is really doing fabulously out there.

Games are free, and you pay for the service and you pay as you go.

So really a good way great way for them to meet their friends and meet people on the Web.

So that's a real growth driver.

Another one, of course, is their purchasing power in the foreign countries are becoming quite good, and as a result they can buy up on their configurations.

One of the best ways to improve your experience to buy up on the configuration is to add a GPU.

So I think those factors are quite important gross drivers for us internationally, irrespective of Vista.

Internationally, Vista is doing well everywhere.

I don't think it's not doing well in China; I frankly think that it's doing very well in China and it will do very well in India as well.

Any consumer will enjoy Vista.

With respect to the application processors, the handheld business, we haven't shipped much application processors.

We haven't shipped any of our AP15 yet.

So all of our current handheld business is from the existing handheld GPU business, which met expectations.

It's not intended to grow, and it's likely to continue to decline.

This quarter I think it did climb.

But we're not expecting that business to grow because we're not out trying to get any more design wins.

We're really focusing now on building our application processor and business where we're going to drive the market.

Do you see a bit of an air pocket, then, between the current gen, the mobile GPU you have now and integrated app processor?

No, not really, mainly because it's not a significant portion of our business right now.

Sidney Ho, Merrill Lynch.

Thanks for taking my question.

It's a question on gross margin.

You are at 46.2% this quarter, and I'm not going to ask for specific guidance.

But if you look forward, say, for the next 12 months, what are some of the biggest [lever section] improved margins, both at the gross and operating margin levels?

Is it more like a product mix between chipsets, GPU, Tesla, or between product families like, say, a G92?

Or was there something else?

We've not increased our gross margins by raising prices once, and the reason for that is because it's hard to grow while you are raising prices.

So we remain very competitive in the marketplace.

Where we have improved all of our gross margins is internal operational improvements.

I'll tell you that I have a lot of work left to do.

So there are quite a bit of internal -- just the things that we can do inside, all within the walls of our company.

We have many, many ideas on improving our gross margins through better operations, better execution.

The mix will affect gross margins, depending on where the growth comes from.

But having said that, our total focus is improving the gross margins on each of the business segments.

Jen-Hsun is right; there's lots of things that we can do better.

So I still see room for gross margin improvement in each of the businesses.

For the Professional Services Group, can you tell us how did average selling price do for the quarter?

Also if you can tell us how much was a contribution from Tesla during the quarter, which I don't expect much, that would be great.

Tesla was very little.

It's just beginning to ship, so it's very little.

We're expecting it to be small, relative to the overall size of the Company, in its first year.

The most important thing about Tesla is targeting it at the type of applications where they could really use the help right away, whether it's oil and gas or imaging, medical imaging, or computational finance.

There are all kinds of applications out there that really benefit from it.

Meanwhile, we're driving CUDA, which is the programming environment, into all of our GPUs, so that we can bring GPU computing to the masses.

So both of those efforts are ongoing, and I frankly think that the progress is extremely high and better than we had expected, by far, by a lot.

As far as the ASPs go, ASPs are in the hundreds of dollars.

It gets to be a complex mix of parts, of products, there.

So let me just say it's hundreds of dollars.

So was it up from last quarter (inaudible)?

No, relatively flat.

Nicholas Aberle, Caris & Co.

Nice quarter, guys.

The question had to do with ASPs.

I know we talked about ASPs here in the most recent quarter.

You guys have done a great job over the last couple of years in improving those.

What is the expectation over time for ASPs?

Do we have a substantial move higher from current levels, based on Tesla?

My general -- the way we think about ASPs here, the way I think about ASPs is we have to make the GPU more valuable.

The more programmable we make it and the more applications take advantage of it, the more people will spend in buying GPUs, because it will enhance their computing experience.

Now, that trend has been moving very aggressively over the last couple of years with all of the applications that I've already talked about and the growth of the gaming market and amazing applications like Google Earth and Microsoft's Virtual Earth 3D.

These type of applications is just continuing to make the GPU more important.

So that's our focus.

My expectation is absolutely the GPU ASP will continue to increase, and the reason for that is because I expect to make it more valuable, and people will see the value and, hopefully, choose to buy a better GPU.

Looking back over the last couple of years, you guys have refreshed Halo product, the enthusiast desktop product, about twice a year, it looks like every fall and spring.

You guys just did 8800 Ultra back in the spring.

Should we be expecting other enthusiast refresh here before the end of the year?

Well, we are at the end of the year.

We decided that the refresh that we do this time was 8800 GT.

This is just a barn-burner refresh, and so we're really, really proud of the 8800 GT, and we're going to put our focus here.

From 8800 GT, you could obviously do SLI, and soon you will be able to do three-way SLI.

So you are going to be able to put a lot of GPU horsepower into your system, starting with the very affordable 8800 GT.

So this is our focus for now.

Intel, back in September, bought a small company called Havok, does acceleration, physics acceleration software.

I just was wondering if you guys are in the market for that IP.

I know you guys have worked with them in the past, and how that may maybe has an effect on the longer-term landscape.

No current plans.

But no commentary regarding how their purchase of Havok affects maybe their IP positioning relative to yours in physics acceleration?

I didn't know that was the question you asked.

You asked me about my plans; I have no plans.

I don't know why they bought Havok.

Physics processing has a long ways to go, and there are so many companies out there.

There's quite a few middleware companies out there that are creating technology in this area.

Many gamers, many game developers incorporate their own physics engine.

So my sense is that there's a lot of invention still left to do in this area.

I'm not sure why they bought that company, to tell you the truth.

It might give them some advantages with respect to Havok, but obviously creates negative synergies everywhere else.

But I don't know why they did, so you're asking, really, the wrong person.

Strategically, I don't know that it has any real bearing on work that we are doing.

Good luck in Q4, guys.

There are no more questions at this time.

I would now like to turn the call back over to Jen-Hsun.

Please go ahead.

Thank you.

NVIDIA's strategy is to be the world leader in visual computing by leading the GPU evolution while creating innovative technologies that enable and inspire amazing applications.

The GPU is increasingly central to our computing experience, and although hundreds of millions of users already enjoy 3D applications and UIs like Vista and Leopard, this is just the tip of the iceberg of the potential of the GPU.

We intend to drive our growth by continuing to extend the GPU to more applications, more platforms and to more people.

Thank you for joining us today.

We look forward to reporting on our results for Q4.

Ladies and gentlemen, that does conclude the conference call for today.

We thank you for your participation.

I'd ask that you please disconnect your lines.