Network Security Internet Technology Development Database Servers Mobile Phone Android Software Apple Software Computer Software News IT Information

In addition to Weibo, there is also WeChat

Please pay attention

WeChat public account

Shulou

The 5G chip has entered the era of the three Kingdoms, who will dominate the ups and downs in the future?

2025-01-18 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > Servers >

Share

Shulou(Shulou.com)06/02 Report--

The world of mobile phone chips has always been bloody. Entering the 5G era, the rivers and lakes have made waves again, and the dispute over the 5G chip has started again. Who wins and loses this time? Who will win the championship?

From the struggle for hegemony among the top five to the tripod, the pattern of rivers and lakes is becoming clearer.

Mobile phone chip generally refers to SoC chip, which integrates baseband chip with CPU, GPU and other application processors. In retrospect, the industry's consistent practice is to conquer the baseband chip first, and then integrate it with the application processor in SoC. The same is true in the 5G era.

As early as February 2019, six chip giants such as Qualcomm, Huawei, MediaTek, Intel, Samsung and Ziguang unveiled 5G baseband chips one after another. In this competition, the process has developed from 10nm to 7nm, and the supporting network capability has developed from only supporting NSA to supporting NSA/SA dual-mode. The integration, performance, energy efficiency and connectivity of the chip are getting stronger and stronger. Until April 2019, when Intel announced its formal withdrawal from the 5G smartphone modem business, the 5G mobile phone chip basically formed the trend of the top five to compete for supremacy.

Since April 2019, with the global operators continue to commercial 5GPY 5G chip competition has entered a white-hot.

In September 2019, Samsung and Huawei released 5G SoC chips, Exynos 980and Kirin 9905G, respectively. But Kirin 9905G uses the latest 7nm EUV technology to locate the flagship. The Exynos 980s use 8nm technology and are considered by the industry to be products that locate mid-range mobile phones. Although Samsung is in a hurry, its performance is lagging behind.

On November 26th, MediaTek released the 5G SoC chip Tianyu 1000, which integrates its own baseband chip Helio M70, positioning the flagship class, taking the lead in all kinds of performance, which has aroused widespread concern in the industry, and is regarded as the overtaking work of MediaTek in the 5G era.

Only a few days later, Qualcomm unveiled two new 5G chip products, Snapdragon 865 and Snapdragon 765. Among them, the middle-end Snapdragon 765 adopts the SoC scheme of integrating 5G baseband, and the flagship Snapdragon 865 adopts the way of external X55 baseband.

At this point, at the end of 2019, the competition pattern of 5G chips has basically formed a tripartite confrontation among Qualcomm, Huawei and MediaTek.

Who wins and who loses? each has its own merits

If you have to compete, compare their exposed basic configurations and performance. Considering that the Kirin 990 was released earlier, while the release time of Tianyu 1000 and Snapdragon 865 is only a few days apart, the two naturally become the object of comparison in the industry.

In the aspect of CPU, Tianji 1000 uses A77 of 4 2.6GHz and A55 of 4 2.0GHz, and Snapdragon 865 uses A77 of 3 2.42GHz of 1 2.84GHz and A77 of 4 1.84GHz. From the perspective of architecture, Tianji 1000 adopts the structure of "4 big and 4 small", while Snapdragon 865 adopts the structure of "1 big and 4 small". It seems that Tianji 1000 is more inclined to the balance of overall performance, while Snapdragon 865 may have better single-core performance because of its higher dominant frequency. The GeekBench running score reported on the Internet also confirmed this conjecture. The mono-core / polynucleus running score of Snapdragon 865 was 4303max 13344, and that of Tianji 1000 was 3811max 13136. Snapdragon 865 was better in single-core performance and similar in multi-core performance.

In terms of supporting frequency band and maximum rate, Tianji 1000 supports below 6GHz band, the highest downlink rate is 4.7Gbps, and the highest uplink rate is 2.5Gbps; Snapdragon 1000 supports below 6GHz and millimeter wave band, the highest downlink rate is 7.5Gbps, and the highest uplink rate is 3Gbps.

Let's first talk about the theoretical rate of 5G NR. In the frequency band below 6GHz, the single carrier bandwidth of 5G NR is 100MHz and supports 4*4MIMO, that is, when the 5G mobile phone supports four antennas, the theoretical peak rate is about 2.34Gbps.

How does Tianji 1000 achieve the maximum downlink speed 4.7Gbps? The reason is the use of dual-carrier aggregation technology, that is, the bandwidth of two 100MHz can be aggregated into 200MHz, and the rate is just doubled.

Dual carrier aggregation technology is very suitable for 5G network deployment of Chinese operators. Take China Mobile as an example, its 5G spectrum resource is 160m bandwidth on 2.6GHz, while the single carrier bandwidth of 5G NR is 100m, which requires the support of dual carrier aggregation technology. Take China Unicom as an example, the construction goal of Unicom's 5G network is to use the 3.5GHz band as the main frequency band for continuous coverage in the urban area (2.3Gps network speed), and the 2.1GHz band to improve 5G coverage and capacity supplement (1Gps network speed). If the mobile phone does not support dual carriers, it can only enjoy the 2.1GHz or 3.5GHz band alone, with the fastest network speed of only 2.3Gbps, and frequent switching between the two bands, while if the phone supports dual carriers Not only can the superposition of 2.1GHz and 3.5GHz network speed be realized through carrier aggregation technology, the highest network speed of 3.3Gbps can be enjoyed, but there is no need to switch, and the 5G network can be kept online all the time. In addition, the dual carrier aggregation technology also supports the co-construction and sharing of Unicom and Telecom 5G, aggregating the 3.5GHz 5G band of 100m bandwidth owned by Unicom and Telecom into 200m, doubling the coverage and rate.

Besides, why can the highest downlink rate of Snapdragon 865 reach 7.5Gbps? The reason is that it supports the millimeter wave band. According to the 5G standard, the single carrier bandwidth of the 5G millimeter wave band can reach 400MHz, and the theoretical maximum rate is naturally much higher than the frequency band below 6GHz.

Why does Snapdragon 865 support millimeter wave band? Because the 5G commercial networks of American operators use the millimeter wave band. However, at present, with the exception of the United States, 5G commercial networks in other parts of the world use less than the 6GHz band, and Qualcomm takes into account the American market.

The biggest difference between the two is that Tianji 1000 adopts 5G SoC scheme, while Snapdragon 865 still uses external baseband. And the plug-in method is different from the X50 plug-in of Snapdragon 855 in the past. Snapdragon 865 also removes the 4G baseband chip from SoC and encapsulates it with 5G baseband in X55, which means that Snapdragon 865 and X55 are sold together.

This surprised the industry. SoC scheme has the advantages of high integration and strong performance, which is the development direction of 5G mobile phone chip, while the plug-in scheme is ineffective and will occupy more mobile phone space, resulting in limited battery capacity, which has been proved in 3G and 4G era. Why didn't Qualcomm adopt SoC scheme on flagship 5G chip?

There are two guesses in the industry. One theory is that because X55 wants to support millimeter wave, some performance and connectivity may be sacrificed after being integrated into 865, and the plug-in solution has the highest performance-to-price ratio; the other is to give priority to meeting Apple's needs. Because Apple has been using the plug-in baseband solution.

Therefore, on the whole, Tianyu 1000 and Snapdragon 865 have their own advantages. Due to the difference in market positioning, Tianji 1000 has the advantage of supporting 5G SoC and excellent performance in less than 6GHz band, while Snapdragon 865 has the advantage of supporting millimeter wave.

From a market point of view, Tianji 1000 meets the needs of the mainstream 5G market, especially in China, where Snapdragon 865 takes into account the American market, but supporting millimeter wave is not very useful for 5G markets such as China and Europe. Until 2021, only the United States needs millimeter wave. According to the data of GSA global 5G operators, only 3 of the 56 will use millimeter waves, and 2 of these three also use less than 6-GHz band technology. This also makes people speculate that there is likely to be a MediaTek version and a Qualcomm version in the future, the MediaTek for the global market and the Qualcomm version for the US market.

The tripod is good for the industry

In fact, from the point of view of the development of the 5G industry, it does not matter who is better than whom. The important thing is that the tripod is good for the 5G industry.

As we all know, 4G changes life, 5G changes the world. 5G network is very powerful, the network speed is 10 times that of 4G, the latency is 10 times faster than 4G, and it can also enable a network to enable digital transformation through end-to-end network slicing. However, reviewing the history of mobile communication, the network has never been the only decisive factor. Chip, terminal and network promote each other to form scale effect, which is the immutable rule of success.

GSM can beat CDMA in the 2G era, thanks to the richness of GSM phones. The unprecedented wonderful mobile Internet life in the 4G era also benefits from the popularity of the dazzling variety of Netcom smartphones. The richer the terminal is, the cheaper it is, the more users connected to the network, and the lower the network cost shared by a single user. The two promote each other, forming a scale effect, thus promoting the development of the mobile industry.

But every time in the past, chips and terminals often lag behind the network for at least 2-3 years, making the industry miserable. For example, the industry's most classic joke took place in the 2G era, when the world's first GSM network was launched in 1991, but without mobile phone support, the executives of German operator Mannesmann complained almost angrily at the second GSM conference that GSM is the abbreviation of "God Send Mobiles". The implication is that I have built the network, and now I can only pray for God to send the phone. The same is true in the 3G and 4G era.

Now entering the 5G era, 5G chips formed a tripod early in the first year of 5G, changing the dominant pattern in the past, which will quickly promote the popularity of 5G mobile phones.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.

Views: 245

*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.

Share To

Servers

Wechat

© 2024 shulou.com SLNews company. All rights reserved.

12
Report