SHANGHAI – China fabless chip company Rockchip is caught between a rock and a hard place.
Just nine months ago, Fuzhou Rockchip Electronics, a developer of apps processor for tablets, looked almost invincible. The company introduced itself at the Consumer Electronics Show in Las Vegas as China’s next rock star. “The sky is the limit,” Rockchip’s proclaimed in a press release.
[Get ready for the
China Fabless Summit 2013, our Spring forum where Chinese startups will showcase their plans for the future of the electronics industry.]
While the sky hasn't exactly fallen on Rockchip since then, the
Android-based media tablet market has gotten far more competitive. By volume, the Android tablet sector has grown to an estimated 80 million units, outpacing the 75 million iPads sold by Apple thus far, according to estimates by Rockchip vice president Feng Chen. But an equal explosion has occurred among suppliers of apps processors for Android-based tablets, with everyone chasing the same market.
Allwinner changes the landscape
Rockchip faces stiff competition from multinational chip companies like Samsung Electronics and Taiwanese behemoth MediaTek. Meanwhile, a host of Chinese chip suppliers are coming after Rockchip, undercutting it on price, features and integration plans.
At the beginning of 2012, the target price of a 7-inch capacitive screen media tablet featuring Cortex-A8 was $99. That price has since dropped to around $65, due largely to Allwinner, a red-hot Chinese fabless company that has flooded the tablet market with its own turnkey system. Speaking of the trend, one industry observer in Shenzhen noted, “It’s a bad news for the whole industry.”
Rockchip's situation vividly illustrates the challenges most Chinese fabless chip companies now face.
During a recent interview with EE Times here, Rockchip’s Chen (right) said, “This is a new world war we’re fighting.” Gazing at the breakfast table, Chen added, “It’s tough. But at least I’ll have a story to tell my grandchild someday.”
Indeed, nearly every apps processor vendor here is in a rough spot because “the time-to-market requirement has gotten much shorter," he noted. "Worse, catching the market rhythm or cycle -- at the right time – has become much harder.”
Rockchip has invested in licensing or purchasing IP cores. In an effort to amortize its investment, the company has been developing as many as six chips a year using those IP cores. “Previously, we may have developed one chip for Tier-One customer,” said Chen.
“But now, as end-product cycles get shorter, we do everything from designing a chip to developing a board and software that goes around the hardware -- literally within a couple of months,” he explained. In March, for example, Rockchip got back from a foundry RK3066, a dual-core Cortex A9 chip with a quad-core Mali-400 GPU. By April, it hustled to showcase sample tablets based on the chip at the Hong Kong Electronics Fair. By May, the company began shipping the new apps processor to its customers' end products.
Betting on Globalfoundries
As the apps processor competition intensifies, the question is how Rockchip can differentiate itself?
Chen said the company may at times have been “a year behind in our prodcut offerings, when compared to apps processors from big guys" like Qualcomm and Samsung. Rockchip’s new strategy is focused on “contracting” the product cycle to fill technology gaps. Hence, it is betting on Globalfoundries’ industry-leading 28-nm SoC platform based on high-k metal gate technology for its upcoming RK31xx chip.
It is amazing to finish a chip design in just 2 months (started a chip in March, showed it in trade show in April and then shipped the samples in May). How much faster can they shorten the design cycle? I have no idea how innovation can take place if there is just little time for chip design to sit down to work, not even think. Personally, I don't think this can help the company to grow healthily and such strategy can't last long.
Rockchip outsources most of the design work to GUC which is a TSMC affliate. My sources told me it usually takes 3 iterations (sometimes 4) to get it right. So the March to May time frame could be the 3rd try, which could be a simple metal change, or from MPW shuttle to full-mask.
Another very informative article - many thanks.
I do have an interesting question - since you are quoting Rockchip's CEO:
The RK31xx is a dual core Cortex A9 product designed for tablets and smartphones. Using Globalfoundries' high-k 28-nm process, “we can reduce cost, gain performance, but most importantly, we will reduce power consumption for tablets,” said Chen
Will the package interconnect for this application processor be Cu-wire wirebond or flip-chip bumps? I am curious how much wirebond for multi-core application processors even in 28nm.
Many thanks in advance if you can find and provide the answer.
I think I am the guilty one here, not articulating this whole "design" process well in the story.
I will correct the story accordingly. Meanwhile, here's an e-mail I just received from Rockchip's Chen. He clarified this for us as follows:
"What I meant was, after we got the chip, we were able to get the chip in end products in less than 2 months."
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.