輝達 (NVDA) 2010 Q2 法說會逐字稿

Good afternoon.

Thank you for holding.

I would now like to turn the call over to Michael Hara, Vice President Investor Relations.

Thank you, sir, you may begin.

Thank you.

Good afternoon, and welcome to Nvidia's conference call for the second quarter fiscal 2010.

With me on the call today from Nvidia, Jen-Hsun Huang, Nvidia's President and Chief Executive Officer; and David White, Chief Financial Officer.

After our prepared remarks, we will open up the call for a question and answer session.

Please limit yourself to one initial question with one follow-up.

Before we begin, I would like to remind you that today's call is being webcast live on Nvidia's Investor Relations website and is also being recorded.

A replay of the conference call will be available via telephone until August 14, 2009, and the webcast will be available for replay until our conference call to discuss our financial results for our third quarter fiscal 2010.

The content of today's conference call is Nvidia's property and cannot be reproduced or retranscribed without our prior written content.

During the course of this call we may make forward-looking statements based on current expectations, these forward-looking statements are subject to a number of significant risks and uncertainties and our actual results may differ materially.

For a discussion of factors that could affect our future financial results and business, please refer to the disclosure in today's earnings release, our Form 10-Q for the fiscal period ended April 26, 2009, and our reports on Form 8-K filed with the Securities and Exchange Commission.

All of our statements are made as of today, August 6, 2009, based on information available to us today and except as required by law, we assume no obligation to update any such statements.

Unless otherwise noted, all references to research market and market share numbers throughout the call come from Mercury Research or John Petty Research.

During this call, we will discuss non-GAAP financial measures.

You can kind a reconciliation of these non-GAAP financial measures to GAAP financial measures in our financial release, which is posted on our website.

With that, I'll turn the call over to Jen-Hsun.

Thanks, Mike.

Good afternoon, everyone.

Thank you for joining us.

We are pleased to report revenue of $777 million, representing sequential growth of 17% from Q1.

We reported a GAAP net loss of $0.19 which included an additional net charge of approximately $119 million to cover costs related to the dye packaging material set issue, which we previously disclosed in July 2008.

Although the number of systems impacted by this remains consistent with our initial estimates, the cost of remediation and repair has been higher than we originally anticipated.

Non-GAAP gross margin, which excludes the additional charge and stock-based compensation, was 36.3%, an improvement of 560 basis points over last quarter.

On a non-GAAP basis, net income was $0.07 per share.

We reduced inventory by another $48 million, which now stands at 51 days on a non-GAAP basis.

We made good progress managing down operating expenses for efficiency projects.

I also want to acknowledge the sacrifices Nvidia employees are making so our Company can continue to invest in the important initiatives we are passionate about.

Strategically, we're focused on two major game-changing ideas.

First, computing is evolving from central processing on to CPU to co-processing on the CPU and GPU.

To enable this new computing paradigm, we invented the CUDA parallel computing architecture that is now shipping in over 100 million GeForce, Ion, Quadro and Tesla GPUs representing a significant installed base for application developers.

After three years of evangelizing, GPU computing has surely reached the tipping point.

CUDA has been adopted in a wide range of applications, in consumer applications nearly every major consumer video application has been or will be accelerated by CUDA.

We estimate there are over 1200 research papers based on CUDA..

We've highlighted 500 of them on CUDAzone.com.

CUDA now accelerates Amber, an important molecular dynamic simulation program used by more than 60,000 researchers in academia and pharmaceutical companies worldwide to accelerate new drug discovery.

CUDA sped up Amber 50 times.

For the financial market Numerics and Compatible announced CUDA as support for their new counterparty risk application, an achievement 18 times speed up.

Numerics is used by approximately 375 financial institutions.

There are broad ranging uses for CUDA including astrophysics, computational biology, and chemistry, fluid dynamic stimulation, electromagnetic interference, CT image reconstruction, seismic analysis, rate tracing, and more.

Another indicator of CUDA adoption is the ramp of our new Tesla GPU for computing business.

There are now more than 700 GPU clusters installed around the world, with new Fortune 500 customers ranging from Schlumberger and Chevron in the energy sector to BNP Paribas in banking.

Starting this fall with the launch of Microsoft's Windows 7 and Apple's new Leopard GPU computing will go mid screed.

In these new operating systems, the GPU will not only be the graphics processor, but also a general purpose parallel processor accessible to any application.

Our second major initiative is mobile and embedded computing.

It is increasingly clear that the combination of rich applications on the Internet and ultra low power computing technology will trigger the next computing revolution.

Rich applications on the Internet can be connected to tiny, yet powerful computers to enable amazing new experiences.

Apple's iPhone, the poster child of such an experience, has revolutionized the phone as we know it.

The Amazon Kindle has revolutionized how we enjoy books.

These are surely the first of many devices to be created that will require small form factor, energy efficient, yet powerful processors.

Some call this cloud computing.

Some call it web 2.0, but all agree it will revolutionize nearly every industry.

We created Tegra to enable the next generation of these connected devices.

One that will deliver the full, high-definition Internet connected experience.

Tegra is the industry's first ultra low power HG processor.

Tegra has been chosen from many mixed generation smartphones, smartbooks, smartpads, media players, TVs and cars.

Let me now comment on each of our four businesses.

GeForce and Ion, Quadro, Tesla, and Tegra.

GeForce and on overall grew a combined 11% quarter to quarter.

GeForce offers unique capabilities that will define the next generation digital computing experiences, like CUDA which makes video and imaging applications like CyberLink, Motion DSP and Nero many times better.

We estimate there are hundreds of millions of users of these applications to create, enhance and enjoy movies, from sharing videos to Facebook, to transferring TV shows onto their mobile devices to be enjoyed later.

With CUDA, video is the new killer consumer application for the GPU.

Second, PhysX is a CUDA software engine that does physical world simulation and brings a new level of dynamics to games.

Just today Ido's highly anticipated Batman Arcum Asylum announced support for PhysX.

Then there's 3D vision stereo.

It's quickly becoming the next big thing for games to movies to photography, there is a groundswell of activity that will burst into the scenes in the near future.

We have announced 26 Ion designs to date.

We currently have nine Ion designs shipping with 17 more in Q3 and Q4.

Ion is about bringing great visual experience to even the smallest, lowest power and most affordable PC.

The market wants a great experience on all PCs, irrespective of size or price.

We have an exciting Ion road map and we look forward to announcing it when ready.

The GPU computing revolution has reached a tipping point.

With the new operating systems the GPU will be a general purpose parallel processor for any application.

In Windows 7, GPU computing interface is called direct compute.

And OpenCL for OS 10 Snow Leopard.

And although the upcoming CPUs and chipsets may include integrated graphics, these graphics controllers lack the general purpose parallel processing capabilities needed by the new operating systems.

Recently, John Petty, a leading industry analyst, forecast the global graphics market to grow nearly 22% in 2010 based in part to the rise of the GPU as a coprocessor.

The report states the continued expansion and development of heterogeneous computing and GPU compute will stimulate growth in 2010, enabled by Apple's and Microsoft's new operating system, new programming capabilities using OpenCL, Direct Compute, and Nvidia's CUDA architecture will remove barriers to the exploitation of the GPU as a serious, economical and powerful coprocessor in all levels of PCs.

Quadro is our professional visualization solution and the standard for professional designers and creators.

Revenue, which has declined nearly 40% due to the broader economy effecting the automotive and other industries, was essentially flat from Q1 and appears to have reached bottom.

Irrespective of current demand, we continue to innovate amazing new capabilities.

This week, we announced [AXE] or application acceleration engines.

Our suite of GPU accelerated software engines that are vital to software companies all over the world including Autodesk, (inaudible) and RTT to name a few.

AXE includes PhysX for accurate PhysX simulation used in CAD to stimulate contact of physical objects.

A multi GPU rendering engine and tools for managing large data sets.

One of the most exciting AXE engines is optics, the world's first interactive ray tracing engine used to generate photorealistic images.

For the first time, designers can study and interact with a photo realistic rendering of their design.

Optics is a must-see to believe new visualization capability.

We are currently engaged with 40 beta sites that are actively using our Optics rate tracing module.

When deployed, each Optics application will be powered by rendering servers with CUDA GPUs.

This week, we also announced Quadroplex SVS, a plug and play solution for scalable visualization.

Quadro SVS will revolutionize the way professionals interact with 3D models and analyze large volumes of data in fields such as energy exploration, architectural design, medical research and consumer packaged goods.

Any application can be distributed seamlessly to run on over four GPUs and displayed across eight displays to create an astounding 32 million pixels of resolution.

We estimate the total addressable market for Quadro SVS is approximately $100 million per year, growing at 10 to 20% per year.

Tesla is the industry's first GPU for high performance computing.

The Tesla solution consists of compilers, tools, libraries, as well as the largest team of parallel computing experts to assist in software development.

Tesla is available as a module, a deskside personal supercomputer, or server for high performance computing clusters.

Tesla achieved its first significant quarter of revenue with approximately $10 million in sales.

Virtually every major OEM, including Cray, Dell, HP, IBM, Lenovo, Silicon Graphics oh excuse me, SGI, Sun and Supermicro now offers Tesla-based solutions.

Tesla computing servers are installed in over 700 sites around the world.

This quarter we supplied our largest single cluster to date, with over 4000 GPUs.

To put this into context, this cluster would be the 12th fastest supercomputer on the top 500 list.

We have over 50 HPC specialized VARs currently selling Tesla today.

We estimate there are approximately 1000 VARs actively involved in the HPC market, which we have yet to engage.

We estimate Tesla to address a $5 billion market opportunity for us over the next three years.

Tegra is our mobile and embedded processor.

We're very excited that our first Tegra project has entered production.

As reported in the press, Tegra is the processor at the heart of Microsoft's new Zune HD.

There are many more innovative devices in the works.

There are 50 active Tegra designs and many more in evaluation.

Currently ranging from smartphones and media players, 35 smartbooks and mobile Internet devices, and a group -- and a growing number of embedded designs for IPTVs in cars.

We estimate that Tegra will address the $4.5 billion market by 2010 and position us at the epicenter of the next personal computing revolution.

Let me now hand over to David to discuss our financial details.

Thanks, Jen-Hsun.

As Jen-Hsun indicated in his opening comments, revenue for the second quarter of fiscal 2010 was $776.5 million, up 17% from 664 million in the prior quarter, and also higher than our estimate of up 5% sequentially.

Our second quarter is historically been a seasonally weak quarter for the industry, so we're pleased with our quarter over quarter growth.

We reported a GAAP net loss of $105.3 million, or $0.19 per share.

This included an additional net charge of $119 million to cover costs related to a weak dye packaging material set that was previously identified and used in certain versions of our previous generation chips.

This compares to a GAAP net loss of $0.37 per share in the prior quarter, which included a one-time charge of $140 million associated with a cash tender offer for the purchase of certain employee stock options.

Excluding this $119 million net charge, as well as stock-based related compensation, we reported non-GAAP net income for the quarter of $37.7 million, or $0.07 per share.

This compares with a $0.09 non-GAAP net loss per share in the prior quarter.

Let me now turn to our individual businesses.

Our GPU business, which includes desktop, notebook and memory represented 48% of our total revenue.

In absolute dollars, it was up 5% quarter over quarter.

The desktop segment of our GPU business was also up approximately 5% quarter over quarter, with continued strength across almost all GeForce product lines, particularly in the higher end performance and enthusiast segments.

The proportion of our desktop revenue sold into the channel was consistent with last quarter, with channel inventories remaining at approximately one month's supply on hand.

The notebook segment of our GPU business was down approximately 5% quarter to quarter.

We lost some share in Q2, as the commercial Montevino refresh platform ramped, we expect to regain share as we ramp into the fall platform with our new 40-nanometer jet PBUs that are in full volume shipments.

Our MCP business, which represented 31% of our total revenue during the quarter, was up approximately 27% quarter over quarter.

Consistent with our first quarter, demand was primarily driven by our chipset products designed for the main stream A&D integrated desktop segment, which was up 55%.

It was also driven by chipset products designed for the Intel-based notebook segment, including Ion, which was up 33%.

Revenue for our professional business, which includes workstation graphics and computing represented 15% of total revenue and was up 10% quarter over quarter.

Substantially all of this increase was attributable to Tesla.

As for Quadro, corporate demand, which comprises a substantial percent of this business appears to have bottomed at the levels we are currently at, with signs we may be in the early stages of an economic recovery.

Our consumer products business, which includes Tegra, as well as embedded entertainment, represented 6% of our total revenue.

It was up in absolute dollar terms almost threefold quarter over quarter.

Tegra and game consoles both saw significant increases.

Looking at the business as a whole, GAAP gross margin for the second quarter was 20.2%.

Excluding the net charges associated with the weak material dye set, as well as stock-based compensation, non-GAAP gross margin was 36.3%.

On a non-GAAP gross margin exceeded our estimates as a result of several factors.

First, we experienced better 55-nanometer yields during the quarter and our mix of 55-nanometer revenue also improved.

Secondly, our consumer business, which carries higher margins because it includes royalties from game consoles, was up as demand resumed following the consumption of excess consumer channel inventories in the first quarter.

And finally, margin in our MCP business was also up as a result of more favorable product mix.

GAAP operating expenses for the second quarter were $267 million.

That's down $25 million from the previous quarter, when you exclude the one-time charge for the stock option tender offer.

This quarter over quarter reduction was primarily related to lower stock compensation expense resulting from the stock option tender offer we completed in Q1.

The full quarter effect of wage reductions implemented in Q1, and lower legal expenses related to ongoing litigation.

As well as numerous other cost reduction initiatives we have undertaken to improve the efficiencies of our operations.

Nvidia's worldwide employment at the end of the second quarter stood at 5506, essentially flat from the first quarter.

Turning to the balance sheet, accounts receivable at the end of the quarter were $352 million, which equated for a DSO of approximately 41 days, an improvement of 1 day from the prior quarter.

Inventories at the end of the quarter were $279 million, down $48 million, or 15% quarter over quarter.

Inventory days at quarter end were 51.

This compares with 64 days at the end of the first quarter and 144 days at the end of the fourth quarter.

Depreciation and amortization expense for the second quarter amounted to approximately $49 million, down significantly -- down slightly from $51 million in the first quarter.

Capital expenditures were $18 million, which as a result of cost reduction efforts were down slightly from the $21 million we spent in the prior quarter.

Accounts payable at the end of the quarter were $276 million, up approximately 50 million from the prior quarter, driven largely by our higher volumes.

In our free cash flow for the second quarter was a positive $117 million.

Cash and cash equivalents, marketable securities at the end of the quarter were approximately $1.47 billion, up approximately $128 million from the first quarter, with the principal contributors being cash earnings and lower inventories.

In closing, let me briefly comment on our outlook for the third quarter of fiscal 2010.

We expect revenue to be up 5 to 7% from the second quarter.

As for gross margin, we expect GAAP gross margin to increase in the range of 36 to 38%.

The primary factors for this improvement are lower overall costs, 40-nanometer volumes, and from higher gross margin products like Tesla and Tegra as they continue to ramp.

We expect GAAP operating expenses to be in the range of $280 million.

This is up from the prior quarter as a result of increased tapeout and prototype activity, as well as increased litigation expenses.

We estimate stock-based compensation expenses in the third quarter to be $23 million and depreciation and amortization and capital expenditures to be approximately flat when compared to second quarter.

Diluted shares for the third quarter are expected to be in the range of 565 million to $570 million.

This concludes our formal remarks.

At this time, we would like to open the call up for questions.

Operator?

(Operator Instructions) Our first question comes from the line of Rick Schafer with Oppenheimer and Company.

Please go ahead.

Yes, hey, thanks, guys.

Nice quarter.

Got a couple questions.

Just one real quick, could you give us a little more like maybe incremental color on the $119 million charge, maybe your confidence level that this sort of puts the bump metal issue to rest, and if so, do you think this is going to make it easier for you guys to win notebook business, retake share going into the back half of the year?

Rick, if you go back a year ago when the Company first recognized this issue and we recorded a charge in the second quarter same time a year ago, that was based upon an estimate at that time that it included a number of factors that the Company, had to estimate didn't have a lot of data on at that point.

At that point, we only had one OEM customer that we had entered into negotiations with.

But here we stand a year later and we've substantially negotiated agreements with almost all of the effected OEMs.

We have more specific information as it relates to what the repair costs will be.

Not only our customers, but other parties that are a part of the ecosystem for supplying those products.

So today, we feel much more confident about our estimate, the range around that estimate.

It is still an estimate.

I won't say it will never change, but we feel pretty good about the data we have today and our ability to estimate it at this point.

As it relates to your second part of your question, if you look at the people that are focused on the bump metal issue it's really a small, very small portion of the Company.

The Company -- the vast majority of the Company really focusing on the future of our business with very little distraction but on (inaudible) certainly it's a distraction for some of our customers and so forth.

It really isn't impacting the Company and our ability to launch new products and for those products to be accepted in the marketplace.

So it really hasn't been much of a head wind for you in the last quarter I guess?

No.

With the -- okay.

Then just a follow-up, can you quantify maybe how much of the second quarter gross margin improvement was related to better mix within GPU, give us a number maybe?

And also, as part of that, how do you guys look at your market share and discreet going forward?

Are you guys sort of more focused on improving gross margin in the back half of the year there?

Or do you expect to see sort of continued pricing pressure as you try to increase share there?

So, on your first question as it relates to GPU margins, we -- we're -- as I indicated in my comments, we have a number of products that cut across a number of different process nodes and this quarter we were able to move more of our business into the 55-nanometer process node.

So that changed the mix.

And probably overall was at least, a point, you might say, in the improvement we saw quarter over quarter.

And as we ramp up, Rick, ramp up our other businesses like Tesla and Tegra, which are inherently higher gross margin products, our gross margin's going to continue to improve.

But the major factor is, as you recalled several quarters ago, we had nearly 150 days worth of inventory and a lot of that inventory was at 65-nanometer and the Company has been really focused on making sure that we work through that 65-nanometer inventory and now we by and large have worked through it.

And as we -- as we're now ramping aggressively 55 and 45-nanometer products and they are all fresh products, the gross margins are going to be much better with that.

Related to increasing margins or increasing share, the answer is yes on both.

I mean we, we believe we can increase margins by driving efficiency in our Company and even though the velocity of this, of the GPU business is incredibly high, we believe that extraordinary excellence can improve margins and we've seen -- you've seen us do that before and on the other hand, we can increase the premium of our products in the marketplace above where it is today.

Almost every single segment, there's a premium on our product and yet we continue to capture a very significant part of the market share.

And the reason for that is because our products have a much better reputation and it's had a history of delivering wonderful experiences and it's just a much more desirable brand.

And so we focus on making sure that the GeForce brand comes with it, unique capabilities that other people don't offer beyond just your basic graphics chip.

You know that GeForce has the parallel computing capability called CUDA, has PhysX processing so that games could be much more dynamic and beautiful, and also this 3D Vision investment that we're making that we've been talking about, it's going to turn into a watershed event towards the end of the year or early next.

You'll know that the entire film industry is moving towards stereo vision and everybody's looking for ways to bring that capability or that experience to the home.

And 3D Vision is a Pioneer in that space.

And from a technology investment perspective, we're just miles and miles ahead.

These kind of innovative ideas are going to continue to help us drive our value to the marketplace and then yet still grow our share.

Great, thanks.

Congrats, guys.

Thanks a lot, Rick.

Our next question comes from the line of Craig Berger with Friedman Billings and Ramsey.

Please go ahead.

Hi, guys.

Thanks for taking the question.

Nice job on the results.

Can you talk about what you're seeing out there on the kind of attach rates for discreet GPUs today in this recessionary environment and where they might be going over the next year?

I -- we've not noticed a shift, if you will, in attach rates per se.

We know we have stronger attach rates in certain segments of the market than others for good reasons.

For example, you heard in the comments that channel inventory is low.

We keep our eyes on channel inventory because a vast majority of our discreet GPUs come from the channel.

So although, although only 50% or maybe 40% of the world's PCs come from Tier 2 providers, system builders and channel integrators, the vast majority of the discreet GPUs come from those 40%.

And the reason for that is because the system builders and the system integrators rely on Nvidia's brand and the additional capabilities that our chip provides, our processor provides to offset the Tier 1 brands that they compete against, whereas the Tier 1s tend to focus on the benefits of their own brand and utilize lower cost components and related to, in this particular case, integrated graphics.

And so our, our attach has historically been higher in system builders than the OEMs -- but that really hasn't shifted or changed from our perspective.

Now, going forward, I think that all of the road signs are starting to become more clear.

We've been evangelizing this idea called GPU computing for sometime.

And it's a very logical and common sensical idea that you ought to use the right tool, or the right processor for the right job.

And, programs, modern applications today have a lot of instruction parallel parts of the software program, but there's an enormous amount of data parallel parts to the program.

In the data parallel processor part of it, which could have been 3D graphics, which we offloaded immediately and historically onto the GPU, but increasingly, it's related to video and images and search and those types of applications in which case we created a parallel processor, a general purpose architecture called CUDA so that applications can offload that.

You're now starting to see more and more people talk about GPU computing.

In fact, I think that there's almost no one that I know of in the computer industry who would tell you now that GPU computing, or in general, called heterogeneous computing is not the future of architecture from industry analysts to Microsoft to Apple computer, to almost anyone developing the future of computer architecture will openly admit that heterogeneous computing is finally here and it's the right approach going forward.

As a follow-up, can you guys talk about where you're at with 40-nanometer, what portion of your output it might be over the next six -- over the next couple of quarters and what the gross margin impacts there are as you pay more for wafers, and potentially get more dye per wafer?

Thank you.

I think the general buzz around the industry is that we are by far the largest producers of 40 nanometer today and I think that that's probably consistent.

We've shipped more and earlier than just about anybody.

And we haven't spoken a lot about our 40-nanometer products because we have, we have fabulous products that we're currently shipping.

So we tend not to preannounce products until the customers are really shipping themselves.

And so we're ramping our 40-nanometer product now.

By Q4, it will represent a very large portion of our revenue.

It certainly won't represent the vast majority of it.

Early next year I think it will continue to grow and my expectation is by about Q3 timeframe, it will represent the vast majority of our products.

Q3 timeframe in next year.

What's the gross margin impact on 40-nanometer versus 55?

Gross margins are higher on 40-nanometer products because they are fresher products and they -- the dye sizes are lower or smaller and so they tend to be, they tend to be more cost effective for us.

Thank you.

Thanks a lot.

Our next question comes from the line of Tim Luke with Barclays Capital.

Please proceed.

Thank you so much.

David, as you look forward and obviously you've seen a significant progression in the gross margin in the July quarter and you're guiding for it to be up again in October, how do you see this moving into the, to the fourth quarter and next year?

Should we think about you having the potential as you move to more 40-nanometer to be able to continue this upward trajectory by several hundred basis points over the next few quarters, or how should we think about that?

And at the same time, could you give us some feel for how we should plan and think about your OpEx going forward as you sort of -- can you hold this 280 level or should we think about it beginning to grow now as the revenue begins to come back?

Thank you.

On the gross margin outlook, there's probably three or so significant pieces of our strategy going forward to increase that.

Certainly as you mention the 40-nanometer ramp will be an important element of that, as Jen-Hsun just commented on a second ago.

The second will be the ramp of Tegra and Tesla, which are two relatively new platforms for the Company.

As they become a larger piece of our mix, their margins today, and we expect to continue, to be above what our corporate average is.

So they will tend to lift the overall margins for the Company.

Quadro, at the same time, carries higher margins than the aggregate for the Company.

And so as the economic climate of our economy and so forth rebounds, and as those two products, Tesla and Tegra ramp, our overall mix is going to be very favorably impacted in our favor.

So, we should see our margins continue to improve, over the foreseeable next four quarters or so.

Will there necessarily be hiccups in the way?

No one can guarantee that there won't be but clearly those are the important elements in our strategy and as they get executed on, we think our margins will continue to make progress.

On the operating expense side, we've indicated previously that our objectives were to get our total operating expense down, in the $265 million-ish range.

Today we ended roughly at that level, but in our guidance, we obviously said it was going to go back up.

So when you look at some of the reasons for that going back up, they are dictated by some of the litigation expenses that are going on, which clearly won't go on forever.

It's dictated by, you might say a convergence of a number of new products, offering, all happening at the same time, which obviously increases some of our expenses in the current quarter.

So I think the, I think we'll continue to see expenses come down from the 280 level, but it will probably just be a little bit slower than what we initially anticipated it will be.

If I may, just two quick follow-ups.

Looked like consumer was up a lot.

Could you give us a feel for what piece of that is going to be Tegra and how should we think about that for the coming fourth quarter?

Then it looked like workstation was fairly flattish.

Is it beginning to show some improvement in, in the -- in workstation, should we begin to anticipate that for the second calendar half of the year and maybe Jen-Hsun, with the introduction of Windows 7, how have you seen that or not impacting the market?

So let me -- I'll just take a couple pieces.

I think they are in general related to, related to gross margin and such.

Our Quadro business appears to have, has normalized or stabilized and certainly the feeling we're getting from our enterprise customers reflects what other people are seeing out in the marketplace that the general economy appears to have, have stabilized some and that the automobile industry, which is a very large consumer of workstations appears to be in better health.

There are other growth markets now for the workstation market.

Medical imaging is a very large part of it.

Medical is -- and medical research benefits a great deal from a lot of recent attention.

We also, we also aren't just waiting for the overall economy to come back, as I've discussed earlier.

We're creating new products that allow people to do new things that weren't impossible before.

This Optics rate tracing engine, realtime rate tracing engine that runs on CUDA GPUs is really a ground breaking idea, ground breaking capability.

We have 40 ground sites going there.

When they deploy, they deploy in relatively large configurations because it's the type of models or the visualization task is quite significant.

And those could -- those 40 beta sites are just as many as we can support.

The number of companies out there who would like to accelerate their time to market through better visualization of their designs is, well, you could imagine how many that would be.

It could literally be every single Company in the world would like to do that.

So this is something that is of great importance and we're dragging that pretty hard.

We also, we also know that high resolution displays and projectors are becoming more affordable than ever.

The Sony 4K projectors, digital projectors are very affordable and people need scalable resolution, scalable visualization solutions to be able to address that and so we created a product called Quadro SVS.

The Quadro SVS virtualizes both the application, as well as the display so you could literally run one application across up to 4 GPUs completely visualized and invisibly.

And then you could take the output of that and literally drive it up to 32 million pixels without the application ever knowing anything about it, so this virtualization technology both at the GPU level, as well as the display level is a ground breaking idea and it's something that we're really excited about.

So we're going to try to -- we believe that innovation drives growth and we can drive growth irrespective of the economy, but the economy certainly has stabilized.

So that's certainly a very important part of our growth factors.

Now, in terms of, in terms of Tesla, the computing market is a multibillion dollars market.

The reason why Tesla has been adopted by literally every single computer company in the world is because they see the same opportunity we see.

People are evaluating and developing software now for Tesla for broad deployment all over the globe and people are looking for ways to improve their computational horsepower per dollar to reduce costs or more importantly, to reduce power.

Tesla servers consumes nearly 20 times less power than a conventional CPU server.

And the reason for that is because of the amount of performance that you get out of it.

When I said earlier doing the right job with the right tool, it's an efficiency statement.

If you do that, it will reduce power, you will reduce costs and you will get things done faster.

So I think Tesla is going to be a very important growth driver for us, as well for next year.

Okay.

And Windows 7, Windows 7, my expectation is that the adoption of GPUs will go up and the adoption of GPUs will go up because very simply that if, if the operating system allows applications to take advantage of the parallel processor in your PC, they will take advantage of it.

And when they take advantage of it, the GPU is the right answer.

Thank you, guys.

Okay.

Thanks a lot.

Our next question comes from the line of JoAnne Feeney with FTN Equity Capital Markets.

Please go ahead.

Hi, good afternoon, folks.

Questions following up on the last one actually.

So, so in terms of the value of the GPU for Windows 7 and others that will take advantage of that, what are you guys thinking, what are you assuming about the future of attach rates or of the chipset business once Intel starts to deploy their (inaudible) which combines the CPU and the GPU?

Yes, in fact, Joann, thanks for the question.

It's probably the greatest opportunity I've seen in the GPU business in the last 10 years.

And the reason for that is because, and we've said this in -- as a high level theme for sometime.

It just had to play out.

The challenge of integrating graphics into a chipset or into a CPU that has very long design cycles is that by definition, you'll have to integrate many of your old technology into a single chip.

And this GPU computing idea that we've been evangelizing for three years, some companies really, really didn't believe it.

They really believed in central processing.

And as a result of that, they underinvested in the graphics technology that goes into these Capellas and to future CPUs with integrated graphics.

And so Windows 7 has a direct compute.

OS 10 has OpenCL.

Linux and others has CUDA.

And people are developing parallel computing applications all over the world.

You see it.

We see it.

And every OEM sees it.

So the only way to really accelerate or even do a reasonable job at those applications is to add a GPU next to it.

So it doesn't really matter where the display comes from.

The display just drives the monitor.

The basic 2D graphics or the basic 3D graphics just drives the monitor.

The parallel processing, the programmable shading and all of those other things will come from another processor and we just call that processor the GPU.

So I think, I think the adoption is in fact going to be quite high when the GPUs integrate basic graphics that are not good enough for the modern operating systems.

Okay.

So what would we have to assume, or what are you guys assuming the attach rate would have to rise to in order to offset the potential loss of the basic chipset business?

What kind of attach rates do you envision arising over the next year or so?

Well, we don't actually expect much loss in our chipset business next year.

We have quite a few exciting projects that we're still working on and, people are pushing us really hard to get more chipsets done.

So we're not expecting the decline of it, but we are expecting adoption to increase for GPUs next year.

Okay.

Can you comment at all, probably not, but can you address at all the issue of the legal dispute over the way you can in the future connect your chipset to the system with Intel arguing that you don't have this particular license that you need?

Well, you didn't say in your question about chipsets, you didn't ask specifically about the future and neither did I comment about that.

And so it's not related.

It's just not related.

No, I--?

We're not -- let me just finish.

We're not necessarily building chipsets for future Intel, Intel buses.

We've not commented anything on that.

And so you're just going to have to wait to see what we come up with.

I think that JoAnne, you've known us for quite a long time now and the one thing that you can probably say about our Company is we're pretty darn clever.

There is a lot of ways to skin the cat and we're going to show you more ways to skin the cat than you can shake at.

And so, so you just have to wait.

Unfortunately, this -- I can't give you the punch line and you'll just have to wait for the big surprise.

Okay?

Okay, okay.

We'll wait.

All right.

Thank you.

You're welcome.

Our next question comes from the line of Patrick Wang with Wedbush Morgan.

Please go ahead.

Yes, thanks for the question, and great job on the quarter.

First, just on MCP for chipset, I know you guys, you guys had some good, some positive trends for AMD, chipsets there.

And then Ion.

I was just hoping you could talk in a little bit more detail about how we should think about those kind of going forward to next quarter or two?

Well, we have, we have Ion projects that are still coming to market and ramping into production.

So my sense is that Ion will still be a growth business for us.

And so we're just going to have to wait and see how the end market adoptions are.

We have quite a few projects that we're still ramping.

The chipset business, the definition of a chipset is just the companion processor or companion chip to the CPU.

And we have a lot of companion chips to the CPU that enhances the overall experience.

Our focus is really very simple.

To enhance the computing experience over the baseline capabilities that Intel or AMD provides.

And so if I generalize the statement like that, the smart people will come up with all kinds of great ideas and we have some really exciting things that we're doing for next year.

Okay.

Got you.

But if we think about it, you guys have clearly seen strong orders just on some of the AMD shipments into Asia, is that something we should expect to kind of continue, or is Ion going to be able to more than offset that type of a decline?

Yes, it's actually kind of interesting.

In a way, we only do well on AMD chipsets when AMD's low end CPUs or main stream CPUs do well.

Right.

So I, our -- part of that question, part of the answer to that question depends on how well AMD does.

So we're rooting for them and we would like them to do well.

On the other hand, we have, we have ion processors that are companions to Intel CPUs.

And so we're rooting for them, too.

And so, we're just rooting for everybody.

Got you.

That's helpful.

Two quick ones.

There's been a lot of concern about desktop pricing intensifying in the back half of the year here.

Just curious how you're thinking about that, what's your sense in terms of how that's going to track and potentially the impact to margins?

Well, we're always going to compete for the business out there and we follow two basic, maybe three basic rules.

One is that -- we always look for ways to innovate.

The competition has a basic graphics chip and they will sell it at the lowest possible price to gain share.

We will tend to take a slightly different strategy all the time.

We're always looking to innovate and create new value that we can bring to the users and in exchange hope that they would reward us with a higher premium.

The GeForce brand across the board has a higher premium almost at every single segment and we hope that they will recognize the benefits of the work that we've done with 3D stereo and CUDA and PhysX so that they will continue to reward us for the value that we bring.

On the other hand, we'll fight hard for business and, we've had to be competitive out there and we're realistic about market conditions and we're realistic about the competition and so we'll continue to fight for our business and then the last lever is to continue to drive costs down in our Company and do the things that are in the control of our Company to improve on margins.

I see.

Thank you.

Patrick, we have guys waiting.

Can we--?

Okay.

Great.

Thank you very much.

Thanks.

Our next question comes from the line of Suji De Silva with Kaufman Brothers.

Please go ahead.

Hi, guys.

Nice job on the quarter.

Hi.

On the professional segment, are you getting the sense that the orders were the ones that were embargoed for a while and now coming back?

Is there any sense there from talking to the sales folks?

What -- Suji, what does embargoed mean?

It just means they put them off basically because of the economic downturn and they were delayed now.

Is that the sense of where these orders are coming from?

It must be.

It's got to be.

My sense is that the requirement for design tools to create products to solve problems and search to discover these signs I think that those -- the demand didn't perish.

I'm certain we didn't lose any market share, and so, so it must be, because demand was held up.

Spending was held up because of cost containment reasons.

We're starting to feel the benefits and some of the early signs of the stimulus program kicking in.

And so I think, I think whether it's an industry or in science, in the academia, we're starting to feel much, much more positive energy relating to upgrades and new machines.

But we're also, Suji, like I said earlier, we're going to innovate our way out of it either way.

And if the market comes back, we'll just get the double whammy, and I can't wait until Quadro goes back to our historical levels of 200 million, $250 million a quarter.

And if we add to that all of the new capabilities that we're creating, as well as Tesla, I think our enterprise business could be really, really significant and so, but we're just going to focus on innovation and when the purchasing power comes back, they are going to have a lot of new things to buy.

One other quick question, the Montevino share that you lost and try to recapture, can you kind of remind us the dynamics of that share shift and why you think that won't necessarily recur going forward?

Thanks.

Well, the notebook business is very competitive and it's a combination of a lot of things.

I forget exactly what the reason was when we, when we lost some share.

Sometimes we don't have the appetite to go for the prices that are needed to win.

Sometimes the competition just has a better offering or sometimes the competition has already designed in and so it's just easier to transition to the new platform.

So there's a variety of different reasons that -- there's only one way to win and about 10,000 ways to lose.

So looking forward, we know what, we know what sockets we've already won and the ramp of 40-nanometers will play a big role and hopefully we come back.

We're looking forward to that.

I think going forward, we'll just have to invent new ideas and new capabilities that really delight our customers in breaking ground and we will win more customers as a result of that.

Thank you, guys.

Thanks.

Our next question comes from the line of David Wu with Global Crown.

Please go ahead.

Yes, good afternoon.

Can you talk a little bit about your guidance in terms of revenues.

I know you had some pretty good seasonal demand, better than seasonal demand in Q2, but Q3 happens to be your -- if I recall, seasonally best quarter of the year and with 40-nanometer-based products coming on stream, why would the sequential growth be such, be such low levels admittedly from a higher base for FY -- the third quarter of this fiscal year?

And I have a quick question follow-on, too.

Well, I think, David, we have -- we're not -- we're not particularly -- we don't have any particular greater insight into the economy than you do, but I think that the notion of seasonality really needs to be challenged or reset with the new economy.

Thinking about the previous season and these seasons, just makes no sense.

It's a new world out there now and so we're just looking at business every single day and we guide to what we think is the most prudent and the best of our ability and it is what it is.

Compared to what some other companies have guided, it appears to be quite good and so we're just going to have to play it out and see how it turns out.

Okay.

This week has been a (Sigrab) week and I understand that our friends at Intel has been talking about Larrabee.

Are there any things that they have revealed to the world I guess this week that is a major -- meaningful surprise or departure from what your understanding was prior to the Sigrab this week?

This is the third Sigrab that they have talked about Larrabee and we paid attention to the last two.

I have to admit that I didn't pay much attention this time.

Not because I don't care; we were just really busy.

So if you hear anything, let me know.

I've not heard anything through the grapevines.

The important thing is that they now recognize that parallel computing is too important not to invest in.

And it's also clear that parallel computing is very hard, otherwise it wouldn't have taken so long to build it.

And to efficiently autoparallelize applications and make it easy for programmers to achieve the level of parallelism that is inherently possible in the silicon is really, really hard work and it's something that we've spent the last 15 years of our Company and for many of us, or many of the industry, many of the experts in our Company some 30 years of their careers working on this problem.

And so it's not something that just because somebody wants to throw a bunch of CPUs together is going to make it happen.

So I think one, they recognize the importance of it.

Two, it apparently is very hard.

And then, three, I can't wait to see Larrabee like anybody else.

After all this talk, I'm just dying to see it.

And so, when I see it, I'll, I'll be -- you'll be the first to know.

Okay.

Just one quick, sneak one quick one in.

What is the tax rate we should be using for this, the rest of this fiscal year or next year?

On a pro forma basis?

Probably this year somewhere in the 4 or 5% range.

Next year, it's anyone's guess, depending upon what happens with the tax legislation.

But I still think it would still be in the 10% type of range.

Okay.

Thank you.

Thanks, David.

Thanks, David.

Our next question comes from the line of Shawn Webster with JPMorgan Securities.

Please go ahead.

Yes, thanks for taking my question.

Can you talk a little bit about the, I guess the overall demand environment.

You laid some context that the economy is still a little bit uncertain.

That colored your outlook for Q3, but based on your order trends, what segments do you think will be the strongest and the weakest and can you give us some flavor on what units and pricing for your GPU business did during the quarter?

I'll let David take the second part and I'll just take the first part real quick because that one is easy.

We're seeing strength right now pretty much across the board.

David?

Shawn, can you repeat the second piece of it?

It has to do with ASPs and units.

Yes, units and pricing.

So they are both obviously up, quarter over quarter, and if you look from an outcome standpoint, we're anticipating them being up again as well.

From an ASP standpoint, it's kind of -- our ASPs are kind of a -- greatly influenced by the mix of the product that we're selling and so we tend to, we tend to have product shifts that tend to maintain ASP in each of the three main stream performance enthusiast categories we try and support.

So I don't know if there's necessarily a strong indicator directionally one way or the other as it relates to ASPs in the GPU segment.

Okay, but the ASPs were up, you said?

When you look at it from a mix standpoint, mix factors into that, so when you look at it by those three segments, yes, they are up in two of those segments and modestly down in the third.

Okay, and then my last one, did you have any benefit from the sale of the inventory you wrote down in prior quarters?

No, no, immaterial.

Okay.

Thank you very much.

There are no more questions at this time.

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

Please go ahead.

Thank you all for joining us today.

We look forward to reporting our results for Q3.

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

We thank you all for your participation.

We ask that you please disconnect your lines.

Have a great day, everyone.