LONDON -- Apple should be looking at acquiring either ST-Ericsson or Renesas Mobile for the sake of their LTE/3G/2G modem technologies, according to Will Strauss, president and principal analyst with Forward Concepts.
Both are commercially distressed companies with parents – STMicroelectronics and Renesas Electronics, respectively – who need to make radical changes to try and stem losses and turn larger semiconductor businesses around. Both have excellent European modem technology but are struggling to achieve full value for that technology through a lack of significant design wins. Which of course leads back to the great success of relatively few players in the mobile device market, Samsung, Apple, Qualcomm.
In his latest newsletter Strauss makes the point that the recent news that Samsung has developed an internal LTE/3G capability which it has deployed alongside a quad-core Exynos application processor, should be a source of concern for Apple. The CMC221S LTE/3G/2G modem is shipping in Samsung Galaxy S3 smartphones. Strauss expects the Via Telecom CDMA modem IC to carry on being included in Galaxy S3 phones for some networks but said that he expects the Samsung modem will be integrated with the Exynos application processor either late in 2013 or early in 2014.
Like Samsung, Apple designs its own applications processors based on the ARM architecture – the A5, A6 and A6X – but fabless Apple gets them manufactured by Samsung. And Strauss observes that Apple has no in-house modem capability. In the past Apple has used 3G modems from Infineon then Intel and more recently has deployed Qualcomm modems.
Samung's potential for integration shows how pressure may be brought to bear on Apple. It might be possible for Apple to license in LTE
capability and include it in a future system-chip but not necessarily the best technology. Integration issues might be much
easier if it was a wholly-owned technology.
"There are few opportunities to acquire an advanced modem design like Samsung's, so Apple has to be exploring possible in-house design (no easy task) or casting eyes on LTE/3G/2G modems from ST-Ericsson or Renesas Mobile," said Strauss, in the newsletter.
There are several TD-LTE-only modem houses that Apple could acquire, but this would only provide a partial solution with more work needed to add FD-LTE plus 3G and 2G capability, he added.
Of course both ST-Ericsson and Renesas Mobile are about a lot more than just LTE modems. They have had to develop complete mobile phone platforms with much software; almost complete phone designs to try and capture phone companies' interest. Which of course is part of the problem. Consumers don't want a lot of mobile phone or tablet computer choice. They want either the i- or Galaxy versions of a seductive mobile dream. So could Apple buy the modem technology from one or other company and leave the rest behind?
I doubt that would suit either STMicroelectronics or Renesas, at least at first blush, but innovative deal making is what makes the world go around. So it could be that one of the last events of 2012 could be a beauty contest between Renesas Mobile and ST-Ericsson with Apple as the admiring swain.
Loss-making ST is close to Apple by way of design wins for MEMS sensors and is expected to make an announcement of a strategic plan for the company on Dec. 10.
There was an old saying "One will never be fired by buying mainframes from IBM". Using the same logic, this saying should be true: "Any Apple employee proposing the acquisition of Renesas Mobile should be sacked". In contrast, ST-E is a much more worthy buy.
I am not saying invest in, I am saying acquire, at least the modem capability.
They can continue to use Qualcomm basebands until such times as they have a better baseband and/or see an over-riding benefit from integrating their own baseband with applications processor.
The difference in graphics is that because Apple is dealing with an IP licensor it is already able to integrate graphics with CPU monolithically.
Don't forget that conversion from feature phone to smartphone is not only the fact of handset makers. When 3G/LTE coverage is not there or data subscription too expensive, feature phone stays the preferred choice.
One of the reason for Samsung's very high growth rate is the conversion of their 100s of millions feature phones to smart phones. And guess who is among the biggest beneficiaries of that growth? Very soon Samsung will likely have predictable new headaches -- China's OEMs on global scene.
"Universal" LTE capability is technologically quite complex and Qualcomm is probably generations ahead of its competitors... But certainly - one can speculate and hypothesize ...
BTW, unfortunatelly this fetaure phone conversion is NOT the case for Nokia
That may be so, for now.
But then you are only as good as the competition and paying somebody else for it.
And will Qualcomm offer to license the IP to Apple for integration in an SoC for 2014?
Maybe Apple has to seriously think about what to buy. WIth processor alone can't really beat Samsung well in the mobile market (not phone but the whole mobile computing). Buying baseband chip or special RF frontend are good direction. I'm not sure if it is really smart to buy either ST or Renesas but it seems there are really not many choices. Apple is losing quite much now so they really need to react much faster if they don't want to end up like Nokia.
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.