Three key factors that determine the success or failure of the 2015 mobile chip war

The MWC2015 World Mobile Communications Exhibition opened in Spain. Various manufacturers released new products one after another. Samsung launched its new flagship phone Galaxy S6, and Huawei launched the Android Wear smart watch. It was very lively.

While mobile phone manufacturers are busy, chip manufacturers are also busy. MediaTek officially announced its new processor brand Helio before the exhibition, and its heavyweight products have made a comeback in the high-end market; Samsung Exynos 7420 debuted, with a record-breaking score; Huawei equipped the long-rumored Kirin 930 on the Honor X2; Qualcomm released the new architecture Snapdragon 820, and mobile phones equipped with Snapdragon 810 have appeared one after another; Intel Sofia plan was implemented, and the smartphone chip was renamed X3 to enter the low-end market.

After 2012, Qualcomm rose to prominence and took up most of the industry's profits. It has held the top spot for many years. Now a new round of battle has begun. Who will be the winner? And what will be the key to success?

1. Performance level of new mobile phone chips

After years of development, mobile processors have become more homogenous. From the perspective of architecture, Samsung's Exynos 7420 and Qualcomm's Snapdragon 810 both use the A57+A53 core structure. In terms of GPU, Samsung uses the new public version of Mali, while Qualcomm uses its own Adreno430.

In terms of technology, Qualcomm, which started the earliest, uses the 20nn process, while Samsung has its own manufacturing process, which uses 14nm. Under the condition of the same core, the one with advanced technology naturally has an advantage, so Samsung's running score is far better than Qualcomm's, making it the fastest mobile phone "chip" at present.

Although Qualcomm released the Snapdragon 820 at MWC2015, claiming that it will also use the 14nm process, but from the current information, the Snapdragon 820 is likely to be a re-creation of an old product, and the experience may not be much better than the Snapdragon 810. Qualcomm, which has always had a performance advantage, will lose to its competitors in 2015.

MTK is more conservative. It says that the current A57 generates too much heat and cannot be used, so it launched the high-frequency A53.

This product, which has just been named Helio X10, is actually MT6795, with 8-core A53 and a main frequency of 2.2GHZ. Its performance may not be better than the current Snapdragon 801, but it has the advantage of lower power consumption.

It is worth mentioning that MTK has a killer weapon in the second half of the year. It has abandoned the A57 generation core and directly used the A72, which is more powerful than Apple's A8X. Samples are already available and mass production will begin in the second half of the year. This progress is likely to be faster than Qualcomm, Samsung and Huawei.

Although Huawei has obtained the 16nm process, it did not dare to produce A57, but chose the low-cost A53 like MTK. Kirin 930 uses eight-core A53 and the GPU uses the previous generation Mali628. Although it is called 930, its performance may not be comparable to 928, but it is just a cost-effective choice.

Intel was originally marginalized in the mobile processor market, but through a huge subsidy of US$7 billion over two years, it has gained a considerable market share in the tablet market. We open the e-commerce page and rank them by sales. Except for Apple and Xiaomi which have brand advantages, the rest are Intel's dual-system tablets.

The X3 series released by Intel this time is actually the bay trail in 2014. The performance is OK, but it is difficult to play well under Android and the running score is not high. This time it is renamed as X3 and entrusted to Rockchip to make the solution to reduce costs. It is ready to start from the low end, targeting the low-end chips of MTK and Qualcomm with A53 core.

2. Forecast of the mobile phone chip war

In terms of performance, Exynos 7420 will be the strongest in the first half of 2015. However, after a year of hard work, Samsung still has not solved the problem of integrating its own baseband chips, and Samsung still uses Qualcomm's baseband.

This means that it is more difficult to make a phone using Samsung's Exynos 7420 chip than Qualcomm's Snapdragon 810. In fact, the performance of Samsung's Exynos chips has been good over the years, but fewer and fewer companies are using them, and only Meizu is still insisting on it.

Huawei's Kirin 930 is actually quite weak in performance. If the new flagship uses 930, it will probably be criticized by the media. Moreover, Huawei has no intention of supplying to other manufacturers, and Huawei still produces and sells it by itself. It would be awkward to use Kirin 930 in high-end products, and it is unnecessary to use it in low-end products (HiSilicon has the quad-core Kirin 6xx). The prospects of this chip are not optimistic.

Snapdragon 810 was criticized recently for many bugs, excessive heat generation, and the 20nm process. However, Qualcomm still responded to the doubts with its products. In the first half of 2015, the only high-performance chip with an integrated baseband chip that is easy to develop is Snapdragon 810, so Qualcomm will still have a good year in 2015.

Sony, LG, HTC, OPPO, vivo, Gionee, Xiaomi, Lenovo, and Coolpad will all continue to be Qualcomm's customers.

As for Snapdragon 820, its performance may not be very high, and it will be launched very late. It will directly face A72, so the prospects are not very optimistic.

MTK will not have any changes in the first half of the year, but will create a big wave in the second half of the year when the A72 and A53 products are launched. In terms of performance, A72 is better than A53. In terms of technology, MTK may have used TSMC's 16nm technology in the second half of 2015, which will have an advantage over Snapdragon 810.

In this way, the flagship products of domestic brands in the second half of 2015 will be considered. Lenovo, OPPO, and vivo are old customers of MTK, and the relationship can be traced back to the era of feature phones. Once the product performance is really excellent, it is entirely possible that these old customers will return from the Qualcomm camp. In the second half of 2015, MTK will get some share from the high-end market dominated by Qualcomm.

Intel's progress is too slow. After acquiring Infineon, it has not yet mastered the integration of baseband and application processors and can only provide chipsets. In addition, X86 cannot perform well under Android. As a result, Intel is not as good as Samsung.

At present, the only smartphones using Intel chips are Lenovo and Asus, old friends in the PC era, and Yuandao and Teclast, new friends in the tablet era. Mainstream manufacturers do not use Intel chips.

From the product perspective, although Intel has advanced technology, the performance of this generation of products is not strong. However, it is right to start with a low price. Judging from the experience of tablets, as long as Intel dares to subsidize, sales can be increased. The prospects of Intel X3 depend on how strong the subsidies are.

3. The key to success in the mobile phone chip war

This round of Android smartphone boom started around 2010, and chip manufacturers have been changing for several rounds. In the comings and goings, we can summarize some rules of this market.

1. To make mobile phone chips, you must integrate the baseband and application processor. Even if you cannot integrate them, you must have your own complete solution. Otherwise, it will not be popular, no matter how strong its performance is.

In fact, early smartphone chip manufacturers, such as TI, nVIDIA, and Samsung, did not have basebands. However, the lack of basebands means that mobile phone development is difficult, the stability is poor, and the strength of mobile phone manufacturers is high.

When Qualcomm, with its baseband, joined the battle, although its early chips did not have performance advantages, Qualcomm quickly became the king thanks to its integrated baseband and simple development.

The latecomer MTK also has a baseband, so MTK quickly rose to become the second largest player, and is showing signs of a comeback this year.

The counterexample is that TI withdrew from the mobile phone chip industry; nVIDIA's graphics performance is unmatched, but few people are interested in it; Samsung's performance is excellent, but it can only entertain itself; Intel can provide a full set of solutions, but cannot provide an integrated SOC, and has also been marginalized.

Qualcomm's performance was not dominant in 2015, but I am still optimistic that it will continue to hold the throne because of its baseband advantage.

2. The high-end market competes on performance, while the low-end market competes on price. Only when the solution price is low can it be really low and will it have a market.

Smartphones now have more performance than they actually have, and low-end chips can also provide a good experience. When the experience is similar, the price is the price of the overall solution, not the chip price.

In this regard, the first to succeed was MTK. As early as the era of feature phones, MTK knew how to provide customers with a full set of software and hardware solutions, the so-called "turnkey", which allowed customers to develop mobile phones at a very low cost, and also caused a wave of copycat phones at the time.

In the era of smart phones, MTK also did a good job, taking into account the cost of peripheral accessories, allowing users to make low-priced mobile phones to compete. It is precisely because of this that MTK has become the second largest in the industry despite its late start.

Another example is Intel. In order to enter the mobile market, Intel offered chips at cost price, but the response was still not good. The reason was that Intel did not solve the problem of peripheral components and control the overall cost. So Intel found Rockchip, an expert in this field.

Rockchip is an expert in reducing costs. It can use cheap peripheral components and simple circuit design to ensure the normal operation of the chip. Intel has established a firm foothold in the tablet market and started to counterattack the smartphone market.

3. Process determines competitiveness, technology is homogenized, and competition turns into a fight for process resources

As smartphone chips accelerate, it is increasingly difficult to develop a non-public version of the core. Apple can design its own core because of its iOS software barrier, while NVIDIA's equally powerful Denver core is not in demand.

Qualcomm's own research and development cannot keep up with the progress, so it has to use ARM's public version A57. However, its own TS2 core (the core of Snapdragon 820) is likely to compete with ARM's powerful A72, and the prospects are not optimistic.

Since everyone uses ARM's public version core, the design is the same, and the final performance difference comes from the process.

The performance disadvantage of Snapdragon 810 is caused by the 20nm process. Samsung Exynos 7420 is powerful because it has its own 14nm process. Huawei does not have the process, but it has started to cooperate with TSMC on 16nm very early.

In the future, the competition for high-performance chips is likely to transform into a competition for process resources. Whoever gets the leading process will have better performance and stronger competitiveness.

Therefore, the deciding factor for a mobile phone chip lies in integration, overall cost control, and process resources. Whoever leads in these three aspects will be the winner.

Qualcomm currently has the upper hand, but the future king may be MTK, Samsung, or HiSilicon.

As a winner of Toutiao's Qingyun Plan and Baijiahao's Bai+ Plan, the 2019 Baidu Digital Author of the Year, the Baijiahao's Most Popular Author in the Technology Field, the 2019 Sogou Technology and Culture Author, and the 2021 Baijiahao Quarterly Influential Creator, he has won many awards, including the 2013 Sohu Best Industry Media Person, the 2015 China New Media Entrepreneurship Competition Beijing Third Place, the 2015 Guangmang Experience Award, the 2015 China New Media Entrepreneurship Competition Finals Third Place, and the 2018 Baidu Dynamic Annual Powerful Celebrity.