AMD Claims To Lead Discrete Mobile GPU Market
There's cheerful news from Sunnyvale today; AMD is claiming to now lead the discrete mobile GPU market with a 36.5 percent market share gain and a whopping 87.2 percent more units sold in the first six months of 2009 compared to the first six months of 2008. At this point, AMD claims to hold 53 percent of the discrete mobile market, and credits the jump in sales to the strong positioning of the ATI Mobility HD 4000 series.
AMD points to design wins with HP (the dv6T), Acer, Samsung, Asus, and MSI, but the HD 4570 also tips up as the default option on Dell's Studio 15 and on HP's ProBook series. "As consumers’ increasingly demand greater visual computing capabilities in their notebooks, the market for discrete graphics is only expected to grow," said Dr. Jon Peddie, founder and principal, Jon Peddie Research. "The increase in volume in the second quarter of this year, and in AMD’s mobile discrete market share, is a testament to consumers’ appetite for visual computing, and manufacturers’ readiness to satisfy it as we move into what has traditionally been the busiest season of the year."
GPU manufacturers refer to the shape of this graph as a "WIN."
That's high praise, but it's unfortunately less clear that the company's market share gains translate readily into hard cash. AMD reported $251 million in revenue from its graphics products division in Q2-2009. That's a gain of 13 percent compared to Q1, but it's just one percent higher than Q2, 2008. Actual unit volumes were up compared to both periods, but falling ASPs hit the company hard.
JPR predicts a strong recovery that will begin in the second half of 2009 and continue into 2010 due to the combination of pent-up demand, new 40nm parts, DirectX 11-class GPUs, and the increasing number of applications that can take advantage of GPU horsepower. If JPR's projection is accurate and AMD continues to execute well, the GPU division's revenue should increase significantly by the end of the fourth quarter.
While a revenue increase is good news for AMD no matter how it arrives, even a 50 percent surge in GPU segment revenue wouldn't bring the company to its break-even point. Using Q2 numbers as a starting point, 50 percent additional revenue would've put the GPU segment at $376.5 million. Such a leap would certainly have cut AMD's net loss for the quarter, but the company would still have been $204.5 million in the red.
In order to reach its targeted Q4 break-even point, AMD will have to do more than simply excel in the GPU industry. According to Dirk Meyer, AMD's manufacturing cost-per-core will fall over the next six months as the company ramps 45nm production; AMD doesn't expect more than 70 percent of its product mix to consist of 45nm parts until Q4. Additional cost-cutting/streamlining will help pull the balance sheet back to black, as will wide availability of the hexa-core Instanbul. It's not beyond the realm of possibility that AMD might yank the latter into the consumer market to help it compete at the higher end against Core i5/i7.
AMD has already established that the Phenom II is ample competition for Intel's Core 2 Quad series, but the CPU manufacturer's ability to earn itself another $200 million (give or take) in microprocessor revenue will depend significantly on how well it stacks up against Lynnfield. That's not a race we can call until Intel's next-generation consumer platform launches, but Sunnyvale clearly has its work cut out for it.
AMD points to design wins with HP (the dv6T), Acer, Samsung, Asus, and MSI, but the HD 4570 also tips up as the default option on Dell's Studio 15 and on HP's ProBook series. "As consumers’ increasingly demand greater visual computing capabilities in their notebooks, the market for discrete graphics is only expected to grow," said Dr. Jon Peddie, founder and principal, Jon Peddie Research. "The increase in volume in the second quarter of this year, and in AMD’s mobile discrete market share, is a testament to consumers’ appetite for visual computing, and manufacturers’ readiness to satisfy it as we move into what has traditionally been the busiest season of the year."
GPU manufacturers refer to the shape of this graph as a "WIN."
JPR predicts a strong recovery that will begin in the second half of 2009 and continue into 2010 due to the combination of pent-up demand, new 40nm parts, DirectX 11-class GPUs, and the increasing number of applications that can take advantage of GPU horsepower. If JPR's projection is accurate and AMD continues to execute well, the GPU division's revenue should increase significantly by the end of the fourth quarter.
While a revenue increase is good news for AMD no matter how it arrives, even a 50 percent surge in GPU segment revenue wouldn't bring the company to its break-even point. Using Q2 numbers as a starting point, 50 percent additional revenue would've put the GPU segment at $376.5 million. Such a leap would certainly have cut AMD's net loss for the quarter, but the company would still have been $204.5 million in the red.
In order to reach its targeted Q4 break-even point, AMD will have to do more than simply excel in the GPU industry. According to Dirk Meyer, AMD's manufacturing cost-per-core will fall over the next six months as the company ramps 45nm production; AMD doesn't expect more than 70 percent of its product mix to consist of 45nm parts until Q4. Additional cost-cutting/streamlining will help pull the balance sheet back to black, as will wide availability of the hexa-core Instanbul. It's not beyond the realm of possibility that AMD might yank the latter into the consumer market to help it compete at the higher end against Core i5/i7.
AMD has already established that the Phenom II is ample competition for Intel's Core 2 Quad series, but the CPU manufacturer's ability to earn itself another $200 million (give or take) in microprocessor revenue will depend significantly on how well it stacks up against Lynnfield. That's not a race we can call until Intel's next-generation consumer platform launches, but Sunnyvale clearly has its work cut out for it.
