Good evening. My name is Lisa, and I'll be your conference operator today. At this time, I would like to welcome everyone to the NVIDIA first quarter earnings call. All lines have been placed on mute to prevent any background noise. After the speaker's remarks, there'll be a question and answer session. If you'd like to ask a question during this time, simply press star, then the number one on your telephone keypad. If you'd like to withdraw your question, please press the pound key. Thank you, and at this time, I'd like to turn the conference up to Mr. Chris Evenden, Senior Director of Investor Relations. You may begin, sir.
Thanks, Lisa. Good afternoon, everyone, and welcome to NVIDIA's conference call for the second quarter of fiscal 2014. With me on the call today from NVIDIA are Jensen Huang, President and Chief Executive Officer, and Karen Burns , Interim Chief Financial Officer. After our prepared remarks, we'll open up the call to a question and answer session. Please limit yourself to one initial question with one follow-up. Before we begin, I'd 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 15th, 2013, and the webcast will be available for replay until our conference call to discuss our financial results for our third quarter of fiscal 2014. The content of today's call is NVIDIA's property and cannot be reproduced or transcribed without our prior written consent.
During the course of this call, we may make forward-looking statements based on the 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 businesses, please refer to the disclosure in today's earnings release, our Form 10-Q for the fiscal period ended April 28th, 2013, and the reports we may file from time to time on Form 8-K filed with the Securities and Exchange Commission. All our statements are made as of today, August the 8th, 2013, based on information available to us as of today. Except as required by law, we assume no obligation to update any such statements. Unless otherwise noted, all references to market research and market share numbers throughout the call come from Mercury Research or Jon Peddie Research.
During this call, we will discuss non-GAAP financial measures. You can find a reconciliation of these non-GAAP financial measures to GAAP in our financial release, which is posted on our website. With that said, let's begin. Our focus on visual computing enabled us to ride out a rough PC market this quarter and is setting us up for growth in new markets. Revenue came in at the middle of our outlook, driven by strength in GPU offset by Tegra. PC gaming performed particularly well, enriching our product mix, growing our share of GPU revenue, and driving margins to another record high. The PC market is evolving as entry-level laptops face pressure from tablets. Yet sales of specialty PCs, like gaming systems and workstations, continue to grow. The disparity reflects how consumers use these different classes of PCs.
Many consumers look to a PC as a general purpose device they can use for browsing email, social media, and video, but much of this can be better served by a tablet. In contrast, gamers are preparing their systems for a strong roster of games coming this fall, including blockbuster franchises such as Call of Duty: Ghosts, Grand Theft Auto V, and Assassin's Creed 4. In a similar vein, for digital artists and designers, the workstation is the tool of their trade. They depend on increasing processing power to deliver results for their clients, and the use of specialty PCs could become even more widespread with the growth of new applications like 3D printing. Open platforms such as PC and Android are outgrowing the walled gardens of the traditional console market. They benefit from more innovation.
Nearly half the developers surveyed at GDC recently were working on PC games compared to only 11% on next gen consoles. PC game revenues are expected to reach around $20 billion annually by 2015, whereas the total for both PlayStations will be less than $10 billion, and Xbox is only around half that. In mobile, games generate 80% of the revenues on Google's marketplace, and two-thirds of the time spent on a tablet is spent playing games. It's this innovation and energy that we're leveraging with SHIELD. It started shipping July 31st, and the critical reception has been excellent. VentureBeat described it as a game geek's dream device. A writer at Forbes said it got him excited about gaming again. CNET called it a high-quality device with stellar performance. SHIELD is an open platform, it can access anything available to an Android tablet.
That means there were literally hundreds of thousands of games available for SHIELD the day it launched. It's a web browser and a way to update Facebook. It excels as a Netflix movie player and as an internet radio or Pandora music player, thanks to the effort we put into the loudspeaker and acoustic design. SHIELD is the best way to play games, Android games, and a key element to our strategy to generate a virtuous circle in open gaming. It links to GeForce on PCs so gamers can play PC games from their couch. One day, it'll be an excellent client for grid gaming. We've paired it with Tegra Zone, a curated collection of games all tested on Tegra, and many of them optimized for Tegra. Tegra Zone has been installed over seven and a half million times already, and that number increases around 10% a quarter.
The model we're building for games is familiar from music and movies. You can buy a song from Amazon and download it to your device or stream it directly from a service like Spotify, or even buy it and host it in the cloud at Google. Similarly, a gamer can download games to SHIELD to play a game online, and soon stream it from GRID. Our content group helps developers differentiate their games on our platforms. This creates synergy across devices. Games optimized for SHIELD will also look great on Tegra 4 devices like the HP SlateBook x2 or the Toshiba Excite Pro, both of which ship this quarter. Games optimized for GeForce will be excellent candidates for grid gaming. In this way, SHIELD, GeForce, and GRID each benefit the other, and gamers benefit from a dynamic ecosystem.
Q2 is a trough quarter for Tegra, but you'll see more Tegra 4 devices launched through the balance of this year. Meanwhile, qualification of our modems, both the standalone baseband i500 and the integrated application processor, Tegra 4i, is proceeding as expected, and we're making steady progress towards certification at a number of carriers. We previewed our next-generation mobile processor, Project Logan, last month. Logan includes our first mobile Kepler GPU, delivering what the technical press described as unambiguous mobile GPU leadership. Licensing our GPU cores, as we announced this quarter, allows us to generate revenue from markets previously inaccessible, such as vertically integrated smartphone OEMs. The mobile Kepler inside Logan has around three times the performance per watt of the iPad 4 GPU, yet performs like a new desktop part.
We showed it running Epic Games' next-generation game engine, Unreal Engine 4, and running Ira, the face demo that as recently as Investor Day in April, was restricted to high-end gaming PCs. Forbes suggested that NVIDIA's Logan could be a mobile graphics disruptor. Gizmodo said that if you get a new phone in 2014 with crazy battery life and amazing on-screen performance, the Kepler GPU will be the reason. Other press were similarly enthusiastic. Last year, we made the decision to pull in Tegra 4i and Logan. This made us late to market with Tegra 4. However, with Project Logan, we're early and well-positioned to capture key design wins. Moving on to our GPU server business, Tesla had an excellent quarter, very nearly its best ever. Notably, this quarter's revenue was driven by application-based deals. That is, customers using off-the-shelf applications are turning to Tesla GPUs to accelerate them.
This validates our continued investment in software. For example, last week's acquisition of PGI, the leading developer of compilers for high-performance computing. Also, since this revenue passes through OEMs, it requires much less sales and support resources from us and offers a more efficient route to scale for that business. In professional visualization, Quadro displayed a strong sequential and year-over-year growth. We launched the Quadro K6000 to complete the lineup of Kepler top to bottom. Prospects look good for future growth. 80% of Quadro sales are driven by manufacturing, and recent PMI and auto manufacturing data has been very encouraging. GRID is showing increased traction in the ecosystem. We announced this quarter that GRID is now supported natively within XenDesktop 7, so enterprises considering Citrix virtualized infrastructures can easily also evaluate GPU acceleration.
We have over 150 proof-of-concept trials underway and a dozen Citrix star partners working on active GRID opportunities. Our focus on gaming, visualization, and compute generated growth in a turbulent market. Leveraging our IP into new markets and applying GPUs to the cloud will accelerate that growth in the future. With that, I'd like to hand over to Karen.
Thanks, Chris. Hello, everyone. I want to highlight some key points before we open up for Q&A. Overall results for the second quarter exceeded our original expectations, driven by strong demand in the high-end segments of our GPU business across desktop, workstation, and server. Gross margins are now in the 56% range, with Q2 being a fourth consecutive record quarter. The performance of our GPU business, and intense focus on cost, delivered EPS of $0.16 on a GAAP basis and $0.23 on a non-GAAP basis. These were $0.04 above street consensus estimates of $0.12 for GAAP and $0.19 for non-GAAP. Revenue for the quarter of $977 million was at the midpoint of our outlook. The GPU business was very strong, with a 9% increase quarter-over-quarter and up 8% over the prior year.
Growth was fueled by the release of new Kepler-based GPUs for high-end and desktop, and with Kepler-based GPUs now available top to bottom for workstation and server. Revenue for the Tegra processor business was down 49% sequentially and 71% year-over-year. We had expected a drop, as discussed last quarter, with the ramp-down of Tegra 3 products and ahead of Tegra 4 shipments. Nonetheless, we are disappointed with the size of the decline. Due to current dynamics in the mobile space, we believe it will be challenging for Tegra revenue to remain flat year-over-year, as originally expected. That said, we anticipate Tegra revenue to be up significantly in Q3 from Q2 as we start to ship Tegra 4-based design wins and SHIELD, our NVIDIA-branded gaming and entertainment portable. We expect further growth in Q4 as more Tegra 4 OEM products come to market.
Our revenue guidance for Q3 reflects these new expectations. We expect revenue to be up approximately 7% from the prior quarter, with stable growth in the GPU business and significant growth in Tegra processor. We again achieved record growth margins this quarter. The large increases you are seeing quarter-over-quarter and year-over-year reflect our strong market position in high-margin segments of our GPU business, particularly in desktop and workstations. For Q3, we expect growth margins to maintain at current levels. OPEX this quarter of $401 million was below our outlook at $7 million below the midpoint. Some of this investment shifted to Q3. For Q3, we expect OPEX of approximately $460 million for GAAP and $418 million for non-GAAP. Our OPEX investments remain tightly aligned to our key growth strategies.
For visual computing, we are extending the technical leadership of our GPUs, growing GPU for accelerated computing, and engaging the market with our GRID GPU servers for enterprise VDI. GRID servers are in production by Cisco, Dell, HP, and IBM. For Tegra SoC, we are leveraging our core GPU investments to bring our visual computing leadership to industries that are being revolutionized by Android and ARM. We can make a contribution to many applications where a great GPU is desired. To date, we have established a proven market position for Tegra that we will build on. Largely with Tegra 2, we generated $360 million in revenue in fiscal 2012 from smartphones and tablets alone. Largely with Tegra 3, we grew revenue 50% to $540 million from these devices in fiscal 2013.
With Tegra 4, we are expanding our reach to other large markets where visual computing matters, such as auto navigation systems, TV set-top boxes, and new desktop form factors like all-in-ones and smart monitors. With non-GAAP OPEX of $797 million in the first half of the year and $418 million expected in Q3, we are tracking at the high end of our full-year expectations of approximately $1.6 billion. We aim to reduce this amount as we remain intensely focused on operating costs. Turning to the balance sheet, our overall position remains very strong. In the first half of the fiscal year, we generated $275 million in cash from operations and $122 million in free cash flow. Cash at the end of the quarter was $2.9 billion, a reduction of $778 million from Q1.
This relates to our capital return allocation to shareholders of $793 million, consisting of $43 million in dividend payments and $750 million for share repurchases. In Q2, we executed an accelerated share repurchase agreement to repurchase $750 million of shares. To date, we have received 40 million shares under this agreement, reducing our outstanding common stock. We expect to receive additional shares when the contract ends in Q3. In addition, year-to-date, we have returned $190 million to shareholders, including $100 million in share repurchases and $90 million in quarterly dividend payments. With this capital allocation strategy, along with our quarterly dividend payments in the second half, we are on track to return capital to shareholders in excess of $1 billion. Thank you. Chris?
Thanks, Karen. That concludes our prepared remarks. We will now take questions. Please limit yourself to one question and one follow-up. Lisa?
Again, that is star one for any questions or comments. Our first question will come from the line of Christopher Rolland.
Hi, guys. How are you? Can you guys talk about initial sales or at least shipments of Shield so far? Is it above, below, or in line with your estimates? Thanks.
Hi, Chris. Yeah, thanks a lot. Shield is doing great. The early reviews are fantastic. You probably read some of the reviews online. People really love that it's a well-made device. It feels exquisite, and the performance is fantastic. It's an open platform, it does all the things that you would expect an Android device to do and more. People are really loving playing with the beta version of our PC streaming and to be able to enjoy GeForce games that looks fantastic on a PC on a small device like that while you're sitting on the couch is pretty amazing. Sales have been great. Everything that we've shipped so far has sold out. We're just starting to ramp production. We've shipped out to our partners only several thousand units, so it's still quite early to tell, but we're expecting to do quite well with Shield.
Okay, great. Also on the Tegra segment, maybe you can give us a little bit more detail there. How should we think about console royalties, I guess, Pachinko, auto versus your regular mobile breakdown, Shield is going to be in there too for next quarter. What might that look like this quarter as well as next? Thank you.
Well, this was our Tegra trough quarter, we had talked about that before, the transition between Tegra 3 and Tegra 4. Because we pulled in two very important projects, Tegra 4i with our integrated 4G modem and Logan, which incorporated our mobile Kepler GPU. As a result of that, Tegra 4 was a little later than our typical cycle. So we had a gap out, we expected Q2 to be the trough. We expect next quarter to be up substantially. As Colette mentioned earlier, however, there were some disappointments this year. Some projects that we worked on, that we had spent a lot of effort on, didn't pan out as we expected. It's relatively well-known now. Nonetheless, it didn't meet our expectations. Overall, I would say that Tegra 4 is going to ramp up very nicely this coming quarter.
There's tablets from ASUS, Toshiba, and HP. There's SHIELD that's ramping up. There are new devices that are going into production that are quite exciting. I think, relative to your other question of auto, we had said earlier that auto is about $100 million for the year or so, roughly. Last year was $50 million, and we're expecting next year to be another 100% growth year. Auto, as you know, design cycles are quite long. The product life is also quite long, but it takes some time to ramp up.
Great. Thanks, guys.
Yeah. Just one last thing. You had mentioned something about game console royalties. We're not accounting for much game console royalties.
Okay, great. Thank you, guys.
Yeah. Thanks a lot.
Hey, Lisa, next question.
Now our next question will come from the line of Ambrish Srivastava with BMO.
BMO.
Hi, this is Gabriel calling in for Ambrish. Thanks for taking my question. Actually, can you give us more details on the licensing initiative for the graphics technology? Also, when should we expect that to translate into revenues? More importantly, help us understand the thinking behind that. What kind of customer pool were you seeing versus the means to diversify yourself?
Sure. Well, we are a specialist in visual computing. This is our core business. Over the years, we've enhanced and extended our business model from just chips to now chips and systems such as GRID, to system components such as GTX cards and Quadro cards and Tesla cards, to now including GPU IP. We're an end-to-end provider of GPU technologies to all of the applications that relate to visual computing. The way that we go to market has everything to do with how the customers would enjoy incorporating our technology into their products. In some customers, they would like to buy our chip and put it on their own motherboard. With some customers, they would like to buy our system components with our software, and we integrate it into their servers or workstations. Some customers would like to buy in retail adding cards of GeForce.
Some customers prefer to build their own chips, their own systems, and take it to market themselves. In those vertically integrated opportunities, we've historically not had an opportunity to serve them with our technology. Now with our mobile Kepler GPU that we revealed at SIGGRAPH, it took us some three years to mobilize the world's most advanced GPU. Now the world's most advanced GPU is in a mobile format. It's the world's first CUDA GPGPU in the marketplace available for mobile, and it takes a huge advance in energy efficiency. Now with this Kepler GPU, we could license our technology in a GPU core format to customers who would like to incorporate it into their own products and address their own vertically integrated market with it. That's basically the idea.
Thanks. As a follow-up, just want to get an idea, as you know, two of the three major game consoles are being launched in the second half of the year. How should we think about the PC gaming business, or what's your view on that?
Well, game consoles are fantastic, I'm looking forward to buying one myself. Today, just like in movies and music, you game on every computing platform. The fastest-growing platforms in the world today are open platforms. PC is the fastest-growing and has now become the largest, some $20 billion worth of global software sales. Obviously, China is very large, but the United States is large as well, Russia and Korea, you name it. PC gaming has become very large. The business models are vibrant, whether it's multiplayer online games, AAA titles, or now free-to-play games. You could see that PC games are just really vibrant. The fastest-growing, that's the largest, the fastest-growing and now the second largest is mobile.
One of the reasons why we really feel that we can make a contribution to mobile is because video games is such a large part of the mobile experience. In fact, it's the largest download of the world's largest digital store, Google Play. Video games is what people spend the most time actually doing on a tablet, and video games and great computer graphics go hand in hand. This is an area where we can add a lot of value. These two platforms are the largest and now the fastest-growing. I don't think there's anything that game consoles will do at this point that would disrupt that. People just have too many PCs, and there are too many people with tablets and mobile devices.
People like to game wherever they happen to be, not just while they're sitting in front of their TV, sitting in front of the family room.
Thanks, Lisa. Next question, please.
Yes, sir. Our next question will come from the line of Blayne Curtis with Barclays.
I wanted to follow up on the Tegra revenue. If you could just talk about the delta. You were looking at flat, now it's expected to be down one. I was wondering if you could just dial in, if you could give any sort of magnitude on how much you expect it to decline now. Then you had some ramps that didn't pan out in the first half, but are you still looking at the same pipeline of stuff ramping in the second half for you? Thanks.
What was the first part of that question? Blayne said it so fast.
If you could dial in any you expect it to decline, do you have any sort of range that you could provide as to how much it would decline?
Relative to last year?
Yes.
Relative to last year. Coming into the year, we had pretty high expectations on one particular platform. There's no sense naming it, but it's a very important platform that also derived from it a lot of design wins. Because this particular platform just didn't do as well as we or frankly, anybody in the industry had hoped, we don't expect as much returns on that investment as we originally hoped. Our guess is that it's going to be down a couple of hundred million dollars or so, maybe somewhere between $200 million-$300 million overall. We don't expect it in the short term to come back. All of the other platforms that we're talking about obviously are not affected by this one particular design win, and all of its associated design wins.
Thanks for that. On the GPU side, you expected growth into Q3. Is that both desktop and notebooks? If you can just talk about the Haswell refresh. Your notebooks were down a little bit in Q2 or the June quarter, or the July quarter. If you could talk about, are you expecting to see notebooks see a bit of a rebound with Haswell? Thanks.
I think that our notebook was down very low single digit year-over-year. It was down single digits from Q2 in the last quarter. It's slightly down. The overall notebook PC marketplace, of course, is not as vibrant as people have hoped. You know that we're not really in the mainstream PC market. We're really in the specialty PC market, whether it's gaming PC or workstation PCs or high-performance computing PCs or premium consumer PCs or PCs that are used for video editing and digital content creation. That's primarily our target. The notebook part of it, of course, the consumer part of it is not vibrant. In our guidance, we're not expecting big growth from it. Our growth from GPU is primarily attributed to gaming, to workstations, the Quadro workstation GPU, and the Tesla GPU.
That's relative to October with that comment?
I'm sorry, Blayne?
My question was, when you look into October, you talked about broad growth. Is both desktops and notebooks growing into October is the question.
Yeah, that's the quarter for Q3. That's where the growth we're expecting. I think that's your question.
Sorry if I'm being difficult. I'm just trying to figure out, I think you're answering it, but when you mentioned all the drivers, Rob, into October, you left out notebooks. I just wanted to clarify, do you also expect notebooks to see a bit of a rebound for October?
We're not expecting much growth from notebook.
Okay. Thank you.
Thanks, Lisa.
Our next question is going to come from the line of Patrick Wang with Evercore.
Hi, Patrick. Thanks for taking my questions. Sticking on GPU, could you talk about the pricing around GPUs now? How will share trend in the second half in your view?
ASPs are up. They've been on an upward trend for some time. That's one of the reasons why also, because they're shifting towards premium products or specialty products like gaming and workstations and high-performance computing, what we call GeForce, Quadro, and Tesla. Because of the growth of those segments, it improves ASP and it improves gross margins. We haven't seen share movements substantially for some time, and we're not expecting much later on this year.
Got it. Thanks for that. Maybe, can you discuss the buyback plan you're looking for in 3Q and 4Q?
Well, we haven't announced any new buyback plan, but we certainly have a commitment to capital return for the long term. The contract that we're under right now does not end until the end of Q3.
Is there any dollar amount we should think about for each quarter?
Well, it was an accelerated share repurchase, we're just waiting for the contract to complete.
Karen, do you want to just tell him what the whole thing is for the year? Maybe he hasn't seen some of the stuff.
The $750 million is what we're going to repurchase, the bulk of it, the majority of it, 40 million shares, we have received, we're expecting an incremental amount in Q3. Effectively, a share repurchase in Q3 as well.
Okay. Will that be complete in Q3, I guess is my question.
That transaction will be complete in Q3. Any further repurchases we'll announce in Q3.
Sounds good. Thank you.
Our next question will come from the line of Kevin Cassidy with Stifel Nicolaus.
Kevin Cassidy.
Dean calling in for Kevin. Thank you very much for taking my call. Could you provide some extra color on the acceptance and progress in the GRID hardware area? We're seeing a number of software announcements. I was wondering how the hardware may be coming along in terms of sale and potentials.
Thanks, Dean. Just as a background for everybody, GRID is our data center GPU server initiative. GRID has several attributes. One, of course, could be used for high-performance computing. One could be used for big data analysis, whether it's image recognition or image processing, or audio processing, or video processing. You know that more and more people are uploading just an enormous amount of multimedia data onto the web. The third initiative is virtualization, enterprise virtualization. One of the most challenging parts of enterprise virtualization is called VDI. These are the major applications of GRID. All of our OEMs have now completed all of their server and data center certifications and are now in production, Cisco, Dell, HP, and IBM.
There are trials going around all over the world, where we see, we have visibility into some of them, and we track about 150 all over the world, in Japan, in Europe, in the U.S., in China. This quarter, we had just started shipping early samples. We'll see how it turns out. But this is our first true enterprise computing application, our first true enterprise computing product. We're quite excited about the prospects.
If you can comment on this, is there primarily interest in customers building their own hardware or in using the NVIDIA design, or how is the interest shaping up in this area?
There are three ways to access GRID. One is buy it from our OEMs. That's Cisco and Dell and HP and IBM, and there's some over almost 30 different server platforms that are now available in the marketplace with NVIDIA GPUs in it. The second way is to build their own data centers, and they would buy GPUs directly from us. These are data centers that are not announced yet, but they're just large data centers that want to build their own data centers. The third way is for small and medium businesses whose customers are single application customers or don't have an IT department, are not big enough to be served by large IT companies, in which case, they'll buy a fully integrated appliance from us. The last, which is the GRID VCA appliance, is not in volume production yet.
It's sampling in North America. The first two are in production.
That's helpful. Thank you very much.
Thanks, Dean.
Our next question will come from the line of Chris Caso with Susquehanna.
Hey, Chris. You there? Why don't we go to the next one?
Okay, we're going to move on to.
Mike, can you hear me? Hello?
Yes. There we go. Everybody good?
Sorry.
Go ahead.
Sorry about that. Just a first question about OpEx and I guess you said that some OpEx was pushed forward into this quarter, but your comments also suggested that the OpEx, I guess you were going to take some actions to reduce the OpEx going forward. Could you expand on that a bit?
Actually, all Karen said was that we're going to target the $1.6 billion that we've said before.
Okay. I guess going forward even, as we look over the longer term, I know you're making a lot of investments in the business at this point. Just some general guidance about the level of OpEx spending we should expect going forward.
I don't think the plan is to reduce OpEx below the current levels. We're investing in three very important initiatives. One is GRID. GRID puts our GPUs in data centers, which powers cloud computing. I can't imagine a more important initiative. The second is Tegra, which is going after the Android and ARM computing revolution. Computing devices are cropping up everywhere. Of course, we know that it's disrupted smartphones. It's in the process of disrupting computing with tablets. Its disruption will propagate out to consumer electronics, to gaming, to TVs, to automobiles, to just about anything with a computing device. Wherever there's a computer, there's an opportunity that some of them would require GPUs. What Tegra is about, that initiative is about, is addressing those opportunities.
Wherever we can address those opportunities, the effort of Tegra, the focus of Tegra spins off because of that intense focus on energy efficiency and targeting mobile devices and small computing devices, it spins off GPU cores that we can now use as core IP. Our investment is really around these three areas, GRID, Tegra, and our core IP initiatives. I think these are three very large and very important growth initiatives that are very important to invest in. We're not expecting to reduce OpEx. What Karen is saying is that we're constantly focused on making sure that there are no waste in the company. We are the largest visual computing company in the world, and our investments are leveraged from visual computing and leveraged from our GPUs. These are still large investments, as you know.
Okay, great. As a follow-up, I just want to make sure that I understood the comments earlier with regard to the Tegra expectations for the year. I think what you said was $200 million-$300 million in Tegra revenues overall. If you could just clarify that, make sure that I got it right.
It's actually backwards. I think last year we had done something along the lines of $750 million in Tegra, in overall Tegra, from smartphones to tablets, to automobile, to embedded entertainment, whether it's WinRT or Android. Okay? All of that was under Tegra, and last year was about $750 million. This year, we're expecting it to be somewhere between $200 million-$300 million less than that. Even though the second half is going to be a big increase from the first half, we're not expecting the increase in the second half to make up for the full year last year. Was that clear, Chris? Was that clear?
Okay, we'll go ahead and move on to our next question, and that is from Betsy Van Hees with Wedbush Securities.
Thank you for taking my question. You guys have done a really great job of growing gross margin quarter-over-quarter, I believe you guided it for 56% this quarter. Are we going to continue to see as we look forward, the continued strides in gross margin, or are we kind of plateauing out here? That's my first question. Thanks.
Hi, Betsy. There are several drivers for gross margins. One is Quadro. Our professional visualization products is quite rich in gross margins because it's mostly a software product. Our value proposition in Quadro is very heavily related to all of the middleware and tools and all of the algorithm technologies that we incorporate into software packages all over the world that are processed by Quadro. Tesla is rich in gross margins because, again, it's very much a computing platform product. It's not a chip product. It's about the programming model, it's about the programming language, it's about compilers and tools and libraries, and all about all of the computational mathematicians that work with our supercomputing centers and research centers around the world to solve the world's most complex science problems. What also drives our gross margins is GRID. GRID, again, is a heavily software product.
The GPU hypervisor, all of the compatibility that's integrated into all of the hypervisors and enterprise infrastructures around the world, the value proposition that it delivers, it's just very much a software product. Of course, in the future, if we were successful with our GPU core IP strategy, that would be a big lift on gross margins. We have products that are very software intensive, very IP intensive, and therefore, very margin rich. We also have OEM products where we have to compete against other chip providers, and our margins are much more compressed there. It just depends on which one of these parts of our businesses grow faster. I'm not exactly sure whether I'd be happier if the gross margins were to be richer and richer, or whether our revenues were to outpace our gross margin growth.
We have the vehicles now from IP to chips to very value rich, software rich system components and systems to be able to drive our growth. Yet we can be singularly focused on visual computing. We can be singularly focused in being the world's best GPU provider and serve these markets that are growing quickly.
Thanks, Jensen. That was very helpful. As we look at the guidance and we look at the GPU business, Tegra, and the licensing, is it fair to say that the GPU business is going to be in line with the midpoint of your guidance, and then you're going to have the significant growth in Tegra, but that's going to be offset because of some of the declines in the video game consoles? Is that how we should be looking at the quarter?
I think that's a good approximation. Our GPU product line basically is midpoint of margins. If you see Quadro growth faster than top line, if you see Tesla growth faster than top line, if you see GRID growth faster than top line, then I would expect our gross margins to increase. If you see Tegra growth faster than top line, then you would see our gross margins come under some pressure. If they both grow at similar rates, then our gross margins would stay flat.
Okay, thanks. I know I said that was my last question, but I do have one more, and that's, I know you guys don't give forward guidance, just the quarter. As we look at the January quarter, what's typical seasonality in the January quarter? I know things have changed a lot in the dynamics of the PC market, but how do you think we should be looking at seasonality for the January quarter?
I think there are seasonality factors, but I think that there are product cycle factors that are greater than the seasonality factor. Although, don't take this as a guidance for the January quarter, but we're just in the process of ramping Tegra 4 into all the devices that have been announced and devices that have not yet been announced. We're ramping SHIELD, and the holiday season, of course, is a factor there, and SHIELD is a brand-new product for us. We're ramping Logan. We were late to market with Tegra 4, and it was a bit of a scramble. I think the decisions were really good. We wanted to bring forward our modem technology so that we can start to engage the market and start the long process of certification. We also wanted to pull forward, very importantly, our mobile Kepler GPU.
It was a three-year endeavor to take the world's most advanced GPU and mobilizing it, and the results are just really quite spectacular. It's kind of amazing to see the graphics that you see in a high-end PC, or even a supercomputer for that matter, running in a little mobile processor. The architecture and the efficiency is so exquisite that the energy efficiency is really, it's a big discontinuity. We've already demonstrated some three times better energy efficiency, and as you know, energy efficiency, if in a constrained energy environment like a mobile device, the most energy efficient processor is also the most high performance processor. I'm super excited about Logan, and now we can get into the marketplace well ahead of the competition.
Those factors, along with the work that we're doing in GPUs, are likely to play a role, and so we'll see how it turns out.
Thanks, Jensen. Really appreciate it. Thank you.
Yeah. Thanks, Betsy.
Now our next question will come from the line of James Schneider with Goldman Sachs.
Good afternoon. Thanks for taking my question. I wanted to ask about the qualification at the carriers for the LTE modem. I think you talked you're still on track for AT&T qualification by the end of this year. I was wondering if you could maybe give us some help to understand whether you might expect that the Tegra 4i product or the LTE modem might ship on other carriers before AT&T?
Let's see. First of all, our modem qualification comes in two different parts, and they're related, but not the same. First is the data service, which is important for all of our modem, excuse me, all of our tablet products. Data modem cert is well ahead of the voice, and we're in the process of doing that as we speak. There's the voice that goes with Tegra 4i. That happens, the certification completes near the end of the year. We're not expecting. The reason why we do AT&T is because it's the most rigorous. If we get through AT&T, you're pretty much set on just about vastly the Western world's open markets. That's the reason why we have focused on AT&T first.
Thanks. That's helpful. Just as a quick follow-up, on the gross margins you saw in Q2, was more of the improvement due to the mix-up of the Quadro and Tesla businesses, or was more of that due to the increase in mix within the desktop GPU business?
All three. Yeah. All three. Yeah. Chris mentioned that Tesla had almost a record quarter, and most of that was from not one singular supercomputing center installation. It was all from open markets and smaller installations and clusters being purchased to run off-the-shelf applications. There's now just a current of business that appears to have developed around the world, and that's really healthy for gross margins. Of course, whenever we do well in high-end gaming or in gaming, it helps our gross margins. You know that GeForce has two parts. There's the GeForce for OEMs, and there's GeForce GTX for gaming. Usually when I say gaming, I just mean GeForce GTX, which happens to be most of the upper half of our GPU business, GeForce business.
That's helpful. Thank you.
Our next question comes from the line of Craig Ellis with B. Riley.
Taking the question. Jensen, just to clarify the question that Chris had asked. You had mentioned that you expect operating expense to stay at current levels. When you refer to current levels, were you referring to the levels that you'll be at with fiscal third quarter guidance or levels that you were at with the result that you had in fiscal 2Q?
This coming quarter, plus or minus.
Annualized, if that's where OpEx is, then you'd be on something closer to a $1.84 billion annual OpEx run rate, correct?
No, that's not correct.
No.
Go ahead, Karen.
I mentioned we're at just under $800 million for the first half, and then if you add the $418 million guidance and then on a path for Q4, if you do some trajectory, small, we're just above $1.6 billion.
Okay.
The classic range of ±2%.
Okay. Got it. Switching gears to the product side, Jensen, a lot of discussion on GRID and SHIELD as they get started to ramp for revenues. When you look out at the ramps that you have planned for those separate products, when do they start becoming material relative to total company revenues? For example, when would either of them start to move into that 3%-5% of company sales range?
I would really love to see GRID be able to contribute at a level like Tesla next year.
And-
Tesla contributes materially. I think that SHIELD is just hard to say. We have high expectations from a personal level, but we didn't set high expectations from a financial level. We think it's just a really special and quite a distinguished product. The more people that are playing with it and touching it and holding it and enjoying it, the more they become really passionate around it. It's a little bit like a gaming PC is to a PC. This is a gaming Android device to an Android device. It's a specialized Android device, just as our GeForce PCs are really a specialized PC for gaming. I think the logic is relatively similar. We can deploy the company's DNA expertise in graphics and gaming into SHIELD. People understand it when NVIDIA brings a gaming product to market.
They've known and grown up with GeForce for many of our. I even meet many engineers at NVIDIA who grew up on NVIDIA's GPUs. They know our gaming brands very well, and they know what we can do in this field. I think it could be a really interesting outcome. We're modest with it right now, and we're just really grateful that everybody loves it so much.
With the growth that you're seeing in the GRID.
That's great.
Yeah.
Can we follow up with you afterwards? We've only got a couple of minutes left, and I'd like to get one more question if possible.
Sure, Chris. Thanks.
Thanks, Craig.
Yes, our next question will come from the line of David Wong with Wells Fargo.
Thanks very much. Your notebook GPU sequential trend for July and October, is this purely a reflection of the overall notebook market, or have you had any significant socket losses that would result in the market numbers showing a falling notebook GPU share for NVIDIA?
Our sequential and annual notebook decline is dramatically slower than the overall notebook decline.
Okay, great. Do you expect Tegra 4 sales might ramp to a peak equal or greater than Tegra 3, or should we expect a lower crest for Tegra 4?
That's a tough question because two factors. One factor is, it's hard to say exactly where SHIELD is going to be, and we're just trying to be modest at the moment. Two, the separation between Tegra 4 and Tegra 3 or Tegra 3 and Tegra 4 is well over a year. The separation between Tegra 4 and Tegra next is, as you know, relatively short because Logan is already sampling. The sacrifice that we made on Tegra 4 to bring in Logan was really a good decision because we wanted to introduce the mobile world to the most advanced GPU in the world and what NVIDIA's core DNA is all about. It's at a time when the market is really craving a great GPU. The next generation, Tegra or Logan, has a lot more than a great GPU in it.
There are many other surprises that we have in store that I'm just super excited about. At the very minimum, it contains the most advanced GPU in the world, binary compatible, architecturally compatible, but at a power envelope that's 100 times lower. It's the most efficient GPU that's ever been built for mobile. Any energy or performance comparison is against anything that's available or around the horizon is just off the charts. We're gonna take Logan to market as fast as we can. Those two factors, SHIELD and Logan, makes your question a little hard to answer.
Great. Thanks very much.
I think the net of it is good news.
Okay, good. Thanks.
Thanks, everyone. We look forward to talking to you next time on our Q3 earnings call. Anyone we haven't caught up with now, we'll catch up with you immediately after the call. Thank you.
This does conclude today's conference call. You may now disconnect.