Deciphering Apple’s next big game?

Apple is no longer the original "Apple". Since the launch of its chip-making activities, it has made great achievements and has successfully ranked among the top ten semiconductor manufacturers in the world in 2019. After the recent shift to self-developed Mac computer chips and its success, Apple's next round of battle has surfaced again.

According to Bloomberg, Apple has begun developing its own internal cellular modem, or baseband, to replace the Qualcomm baseband currently used in iPhones in the future. Not only that, it is rumored that Apple will also charge into the front-end RF field and include it in its "chip map".

Baseband + RF, what does this combination mean? What impact will Apple's move have on the industry?                                            

advance and retreat?

After all, it can be said that Apple has suffered from "baseband" for a long time.

That vigorous lawsuit battle seemed to be still before our eyes. In the end, Apple had to "rely on strength" to lower its posture. After three years of fighting, it chose to compromise and signed a six-year patent licensing and supply agreement with Qualcomm, while paying a large sum of fees.

Only by looking back at this origin can we better understand Apple’s choice.

In the first few generations of iPhones, Apple used Infineon's baseband, but in 2014 it switched to Qualcomm, which had better performance. Infineon lost its financial backer and was acquired by Intel for $1.4 billion. Later, due to antitrust investigations, the relationship between Apple and Qualcomm broke down. In 2017, Apple turned to Intel, but Intel's 5G baseband was delayed in supply and its performance after mass production was unsatisfactory, resulting in iPhone XS and XR being affected by the baseband, resulting in poor signal and unstable problems. For the sake of the overall situation, Apple and Qualcomm reached a settlement in April 2019.

But obviously, Apple has not given up its efforts to get rid of Qualcomm. In July 2019, Apple announced the acquisition of most of Intel's baseband business for US$1 billion. Through the transaction, Apple obtained more than 17,000 patents for wireless technology, and about 2,200 Intel employees will join Apple. In addition, according to Bloomberg, Apple has been "hiring" engineers from Qualcomm for many years to help it develop basebands.

Judging from the past, the difficulty of baseband development has obviously become the "main line" of this tortuous plot. According to reports, the complicated communication standards, coupled with the increasing number of frequency bands, have increased the difficulty and complexity of baseband development. Not only do manufacturers themselves need to be familiar with communication formats, standards, protocols, and technologies, but they must also be compatible with various base stations and frequency bands of major communication equipment manufacturers. In addition, 5G basebands also need to be compatible with 2G/3G/4G networks at the same time, and meet eMBB, mMTCL, and uRLLC, which means that the baseband needs to have greater flexibility to support multi-mode full-band.

In short, it not only tests the process or post-production, but also values ​​the accumulation of long-term experience. This is why it took 8-10 years for self-developed baseband manufacturers such as Huawei and Samsung to catch up with the progress of the first echelon. Now, Apple has to overcome these tangible and intangible barriers by itself.

If the baseband is the Achilles' heel that Apple is determined to break through, then why does it want to take over the RF as well?

Scheming?

This can be traced back to the continuous evolution of communication standards.

With the commercialization of 5G, technologies such as Massive MIMO and multi-band are widely used in antennas, which not only involve base stations and mutual compatibility issues, but also bring the biggest challenge to mobile terminals, which is the coexistence problem. How to reduce the influence and coupling between each other, how to increase the isolation of channels, etc., all put forward new requirements for 5G terminal antennas and system architecture.

After the terminal antenna receives the signal from the base station, it must be processed by the RF front end before being sent to the baseband, and vice versa. Therefore, technology veteran Dai Hui mentioned that if the RF is not done, the baseband will be difficult to do well, and overall performance optimization will be difficult to achieve. The previous iPhone's mobile phone signal was not good, on the one hand, it was a baseband problem, and on the other hand, the RF was not adjusted properly. For Apple, using Qualcomm's baseband is a stopgap measure, and the baseband and RF are inseparable, so it is advisable for Apple to choose to advance in both directions.

As 5G commercialization moves towards 2.0, the integration of mobile phone baseband, AP and even RF front-end will become a "compulsory course" to test the skills. Apple has been in the AP field for many years, and it is reasonable to strengthen research in this area. Industry analysts also believe that if Apple can successfully integrate the baseband into the SoC, then not only the iPhone, but also various wearable devices such as Apple Watch and AirPods will have the opportunity to move towards independent operation and achieve direct networking instead of relying on the iPhone.

From the perspective of competitors, since Huawei was banned and Honor became independent, Apple's opponents may turn to Samsung and a number of domestic mobile phone manufacturers linked by Qualcomm. Samsung has many years of experience in deploying base stations, and has also established a first-mover advantage in the fields of mobile phone baseband and RF. It should be noted that the system solution provided by Qualcomm based on the baseband includes baseband, RF front-end, receiver and antenna elements, so as to achieve the best performance in terms of power consumption, performance and cost benchmarks. Apple uses Qualcomm as a 5G baseband supplier, but does not use Qualcomm's bundled RF360 as an RF front-end supplier. Obviously, if Apple wants to continue to optimize terminal performance, it is not inconsistent to enter the RF field.

Not only that, judging from the trend, the current industry giants are all planning for 6G, which also means that mobile phone design will become more complex. RF chips are responsible for the quality of wireless communication and are related to user experience. Dai Hui mentioned that in order to maintain its advantage in the future communication market, Apple needs to increase investment in baseband and RF fields, enhance its autonomy, and further occupy the high ground of the integrated software and hardware ecosystem.

In terms of the path, Dai Hui believes that Apple may customize RF from suppliers such as Qorvo and Skyworks in advance, but will integrate its own core technology into it, thus differentiating itself from other companies. Then it will slowly accumulate experience and further open up the situation.

Change?

As the saying goes, "a single move affects the entire body." Apple relies on its huge size, and any action it takes will have a profound impact on the supply chain system.

Obviously, the direct impact is on baseband and RF suppliers. Qiao Shan (pseudonym), a representative of a domestic RF manufacturer, interpreted this and said that Apple's mobile phone baseband uses self-developed chips, which has the most direct impact on Qualcomm. Qualcomm will lose this important customer, which will have a certain impact on the company's revenue. In addition, Apple's RF is mainly sourced from international manufacturers such as Broadcom and Qorvo, which will inevitably cast a shadow on the development of these manufacturers.

Where there are losers, there are winners. Johnson further analyzed that it is highly likely that Apple's self-developed baseband will be taped out at TSMC, which will be good for TSMC. As for self-developed RF chips, they will basically be taped out at Win Semiconductors in Taiwan. Recently, it was reported that Win Semiconductors invested NT$85 billion to expand production capacity, which is likely related to this. Of course, the demand for GaAs process brought by 5G is the main driving factor.

In addition to the impact on related manufacturers, Chen Yuenan, senior analyst at Jiwei Consulting, also believes that Apple's self-developed team's future collection of technology in the industry, such as mergers and acquisitions, poaching teams, etc., will also cause fluctuations in the overall industry.

The demonstration effect of Apple's move on other terminal manufacturers is also worth noting. Chen Yuenan analyzed that industrial chain integration is definitely the ultimate form. Samsung will further collect its core key businesses. At present, Vivo and OPPO are also investing heavily in R&D. However, they may not be able to make many core chips now. In the future, if conditions permit, they should continue to expand.

As the saying goes, "blessings and misfortunes go hand in hand." Although entering Apple's supply chain is a good thing, over-reliance on Apple will also bring great risks to the company's business. The previous experiences of Imagination, Intel and other players have also proved this point. Learn from the lessons and never regard this business as a permanent business or a pertinent one.

"For independent RF manufacturers, the first is to strengthen R&D to maintain technological leadership, the second is to abandon some non-core businesses, and the third is to find suitable targets for mergers and acquisitions, improve their own market voice, and enhance their competitiveness," Chen Yuenan finally suggested.