Analyst Conference Summary

conference date: November 10, 2016 @ 2:00 PM Pacific Time
for quarter ending: October 30, 2016 (Q3, third quarter fiscal 2017)

But I own competitor AMD
Forward-looking statements

Overview: "Breakout quarter," records all round, as VR, datacenter acceleration and automobile applications add to the traditional gaming revenue.

Basic data (GAAP) :

Revenues were $2.00 billion, up 40% sequentially from $1.43 billion, and up 54% from $1.31 billion in the year-earlier quarter.

Net income was $542 million, up 114% sequentially from $253 million and up 120% from $246 million year-earlier.

EPS (earnings per share) were $0.83, up 108% sequentially from $0.40, and up 89% the $0.44 year-earlier.

Guidance:

Q4 fiscal 2017, ending January 30, 2017, revenue expected within 2% of $2.10 billion. Gross margin 59.0% GAAP and 59.2% non-GAAP. Operating expense $572 million GAAP, $500 million non-GAAP. Tax rate 19% to 21%. Capital expense $45 to $55 million.

Conference Highlights:

"We had a breakout quarter - record revenue, record margins and record earnings were driven by strength across all product lines," said Jen-Hsun Huang, CEO of Nvidia. "Our new Pascal GPUs are fully ramped and enjoying great success in gaming, VR, self-driving cars and datacenter AI computing. We have invested years of work and billions of dollars to advance deep learning. Our GPU deep learning platform runs every AI framework, and is available in cloud services from Amazon, IBM, Microsoft and Alibaba, and in servers from every OEM. GPU deep learning has sparked a wave of innovations that will usher in the next era of computing."

The dividend was raised to $0.14 per share. To be paid on December 19 to shareholders of record as of November 28.

GPU segment revenue was $1.7 billion, up % sequentially from $1.20 billion, and up 53% y/y.

Tegra segment revenue was $241 million, up 45% sequentially from $166 million and up 87% from year-earlier. Tegra Xavier will be used for self-driving cars.

In addition to the segments, NVDA reports revenue by a number of market platforms:

Gaming revenue was $1.24 billion, up 59% sequentially from $781 million, and up 63% from year-earlier. Announced won the Nintendo Switch home gaming slot. Extended the Pascal line. Demand for GeForce chips was strong in both desktop and notebook. Gamers are upgrading GPU to take advantage of fall titles.

Professional visualization (Quadro) revenue was $207 million, down 3% sequentially from $214 million, and up 9% y/y.

Datacenter revenue was $240 million, up 59% sequentially from $151 million, and nearly tripled from year-earlier. Most major cloud services are deploying Nvidia GPUs. Broad enthusiasm for AI applications. GRID graphics also showed strong growth.

Automotive revenue was $127 million, up 7% sequentially from $119 million, and up 61% y/y. Will be used for all Tesla vehicles. Working on cloud-to-car mapping systems.

$186 million OEM and IP business, up 14% sequentially from $163 million and down 4% from year-earlier.

Non-GAAP numbers: Net income $570 million, up 82% sequentially from $313 million, and up 124% from $255 million year-earlier. EPS $0.94, up 77% sequentially from $0.53, and up 104% from $0.46 year-earlier. 59.2% gross margin. Excludes $ million in stock-based compensation and other non-cash and one-time items.

Cash and equivalents balance was $6.67 billion, up sequentially from $4.88 billion. $ million cash from operations. $ million was used for capital expenditures. $ million free cash flow. Repurchased shares using $ million. $ million was used for dividends. Long-term debt $2.0 billion.

GAAP cost of goods sold was $821 million, leaving gross profit of $1.18 billion. Operating expenses of $544 million consisted of $373 million for R&D and $171 million for Sales, General and Administrative expense. Leaving operating income of $639 million. Interest and other expense $18 million. Income tax expense $79 million.

Q&A:

Datacenter rapid growth causes? "GPU computing is more important than ever." Drivers are graphics virtualization; computational scientific; deep learning. We transformed our company from graphics processing to general purpose processing, and from a chip company to a platform company. The software stack we provide is complicated, it requires an ecosystem including GRID and CUDA, as well as application libraries for GPU acceleration. It also saves energy and space in the datacenter.

Datacenter lumpiness in the past, looking forward? There are many variables that influence the business. At first we served supercomputing centers. Then workstations. Then hyperscale datacenters. Now they are using P40 and P4 for deep learning. So the business has multiple trajectories, corresponding to multiple applications.

GPUs have reached a tipping point, it is available everywhere, from notebook computers to the cloud.

Expounded on increased use of AI, resulting in increased use of GPUs.

PlayStation and XBox announced 4K versions, which means higher production values, which will drive the adoption of higher-end GPUs. Gaming is now a social phenomena. Market size is growing, and ASPs (average price per unit) are growing too.

Pascal production is fully ramped. But demand is building, so they are ramping in that sense. TSMC yields at 16 nm are fantastic.

Anything you can do with an FPGA, you can do with an ASIC. An FPGA is what you use when the volume is not large or you are not sure what the chip needs to do. Our strategy is about building a computing platform, based on a general purpose parallel processing, and is very flexible. Combine it with a CPU and it is good at both sequential and parallel.

Why is there such a long design-in cycle for datacenter customers like Microsoft? Several things have to happen. Applicatins need to be developed, datacenter software has to handle it, neural network software has to be trained. The Kepler, Maxwell, Pascal series have the same architecture, so future adoption can take place faster.

OpenIcon NVIDIA Analyst Conference Summaries Main Page
NVIDIA Investor Relations page
Openicon NVDA main page

company page: Nvidia

Search

More Analyst Conference Pages: