The Death of X86 Mobile Phones: Falling Behind at the Starting Line

Intel will significantly reduce its investment in the field of mobile SoC (system integrated single chip for mobile phones), and no new SoC products will be seen in the next period of time. However, after widespread media reports, Intel's new senior executive, Venkata "Murthy" Renduchintala, said: We are not exiting the mobile market, but expanding its definition to become synonymous with more than 50 billion devices.

By the way, the new position he took up after leaving Qualcomm is difficult to translate. It is called: "President of Client and Internet of Things Business and System Architecture Group". Yes, this does not sound like a title for a position in charge of mobile devices.

One misstep leads to eternal regret

"I listen to you. I will not go against my best employees." The fate of Intel changed when Jobs said the above sentence.

When the iPhone and iPad were first launched, there were rumors that Apple would use Intel's x86 chip as the processor, and Intel's Moorestown was also said to be designed for the iPhone. However, when Jobs planned to equip the first generation iPad with Intel chips, Tony Fadell, who was still an Apple employee at the time and was regarded as the father of the iPod, said that ARM was much simpler and more power-efficient than x86, and threatened to resign in agitation to prevent Jobs from using x86.

In the end, Intel missed out on iOS, leaving them with a generation of regrets. Intel is the company with the highest level of technology in the chip industry, bar none. The problem is that before the emergence of the smartphone generation, almost all PCs used Intel's x86 processors, which made it easy for them to make money from x86 at the beginning, resulting in the fate of Intel processors being tied to the "x86" ship.

The original sin of x86

x86 is Intel's chip based on the "complex instruction set" (CISC). Compared with their arch-rival, ARM's "reduced instruction set" (RISC), x86 is inherently at a disadvantage in terms of power consumption. Although Intel itself leads the pack in chip technology, they are like a national chess player who gives up his "chariot, horse, and cannon" to play chess with ARM: Intel needs to spend more time, cost, and energy than others to get close to ARM's level in the mobile field - this may be why Fadell strongly opposed the use of x86 chips back then.

In fact, Intel also had an ARM chip line: XScale, but after they sold XScale in 2006, they bet their entire mobile chip business on the "twice the effort for half the result" x86 Atom chip. But after they missed the iPad express train, they were pushed into a deep pit by the Android mobile express train. When Intel was abandoned by the iPad, it had no choice but to turn to the open Android; but a number of mobile phone manufacturers would rather use ARM chips from Samsung, Qualcomm, Texas Instruments and even Nvidia than consider x86 chips.

Falling behind at the starting line

It was not until 2012, when Intel launched the Medfield chip, that the first Intel Inside smartphone was launched on the market: Xolo X900. Then, Intel joined forces with Asus, which had a very low smartphone market share, to launch smartphones with Intel chips. But in 2013, when Intel began to catch up, the number of Android applications had surpassed iOS, but most of these applications were designed based on the ARM instruction set, and theoretically could not run on x86 phones.

However, when Intel was committed to developing smartphones, it actually took into account the compatibility issues of Android and cooperated with Android to launch optimizations (pictured above). They told Inquirer that Intel's x86 mobile phone chips are compatible with 95% of Android applications; however, small applications that generally only use virtual machines are of course no problem, but it becomes very troublesome when running native applications. Many developers pointed out that when running native applications, either developers have to actively support x86, or x86 has to sacrifice performance to convert compatible codes, and the result is that they get "half the result with twice the effort" in development.

In addition to developers working twice as hard for half the results, Intel's development has also become more difficult. Asus, which was once Intel's "good friend", was reported to have reduced its purchase of Intel chips earlier. Analysts pointed out that the reason was that the update speed of Intel chips could not keep up with the speed of Qualcomm and MediaTek. If Intel, which has strong technology, has fallen behind, it is because they have been behind at the starting line for a long time, which makes their chip development speed unable to keep up with others.

When it rains, it pours

Because of this, Intel has done more to promote the mobile market than other chip companies, but the returns are still small. I don't know whether Intel has offered ASUS a discount to use x86 chips, but ASUS's smartphone sales are mediocre, so it has not brought much help to the x86 mobile phone platform.

Therefore, Intel finally even sought a way out from the copycat companies in Shenzhen: According to WSJ, Intel was reduced to providing discounts and even financial assistance to Chinese white-box tablet manufacturers in 2013. These white-box manufacturers said that the discounts Intel provided to them were completely different from what a large company like Intel would provide.

Even if Intel does not have to be cool, it is still difficult to turn the tide. Since 2015, the global economic environment has deteriorated, the mobile phone market is shrinking, and the two chip industry giants Qualcomm and Samsung have been struggling in semiconductor performance due to the slowdown in mobile phone growth. What about Intel? Although the revenue of the data and Internet of Things departments continued to grow significantly, the revenue of the client department fell by 14%.

No x86 phones in the future

Going back to the original statement of the new Intel president, we can notice a very important statement in his speech: After the revolution, values ​​are different. We have changed from the previous SoC orientation to embrace the intelligent and connected world under 50 billion devices. This will bring huge value to our customers and shareholders.

Perhaps, they did not give up the mobile market, but they changed their "previous SoC orientation" and gave up Broxton and SoFIA, which is equivalent to giving up the x86 mobile market in 2016.

Therefore, they have never raised the x86 banner in smart home or VR as they did in the past. Even though we asked Yuneec, an Intel-invested company, whether its drones use x86, they were unwilling to respond directly.

However, as they abandon x86 in the mobile market, the future focus will be on the following two aspects:

Communications technology, 5G;

Sensor Technology:RealSense.

But as the PC market shrinks and smartphones become more important than PCs, does this mean that x86 is going to die? With the support of the server and professional markets, x86 should still be able to survive, but it may gradually fade out in front of ordinary consumers. Will Intel still produce consumer-grade processors in the future?

Intel doesn't want x86 to die, but they have probably already prepared for the worst-case scenario of x86 fading out of the consumer market.

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.