三星(Samsung)近日发布了一款新型 LTE无线芯片，支持 FDD 与 TDD 两种规格，采用28纳米HKMG制程；值得注意的是，三星在新芯片中整合了自家开发的四核心应用处理器，并将之命名为 Exynos ModAP。
也就是说，三星以Exynos品牌整并了旗下的蜂窝组件，模仿高通(Qualcomm)的Snapdragon系列手机芯片；其意图很明显──拿下高通。不过，三星在今日完全由高通主导的LTE市场上能发挥出何种程度的破坏力，仍然有待观察。
三星的举措注定要让高通的芯片远离三星品牌手机与平板设备。新型的Exynos品牌LTE应用处理器组合芯片甚至还未上市，就已经掀起了最强劲的风暴，只是受害者并非高通，而是博通(Broadcom)──该公司在6月初宣布放弃蜂窝基频芯片业务。
无 论我们用哪种方式来切分LTE市场，高通在手机芯片领域的地位都难以动摇；市场研究机构Forward Concepts总裁Will Strauss表示：“就算有一些2014年才进入市场的新竞争者，高通在多模LTE芯片市场的占有率仍超过95%。”另一家市场研究机构 Strategy Analytics的报告则指出，三星2014年第一季在LTE基带芯片市场排名第四，全球市占率(营收)仅1.7%。
2013年多模与单模FDD-LTE 基带芯片供货商市占率
Source：Forward ConceptstBiesmc
三星在LTE调制解调器芯片领域并非新手，该公司已经推出过使用在Galaxy 4与5手机中的多模4G调制解调器；但Forward Concepts表示，三星的活动仅限于韩国市场，在世界其他区域市场销售的手机与平板，该公司都是仰赖竞争对手的LTE调制解调器芯片。
Strauss 表示：“依据区域不同，三星有采用高通的4G调制解调器，同时也采用英特尔(Intel)的4G调制解调器芯片──实际上是用在10吋的Galaxy Tab 3平板设备；此外在7吋的Galaxy Tab 3平板设备则是采用瑞萨(Renesas)/博通的调制解调器芯片。”不过现在这种状况将会改变。
对博通产生的剧烈冲击
对 于三星推出的LTE/调制解调器组合芯片，以及LTE-Advanced调制解调器芯片──两者都是采用CEVA的DSP──产业观察家Strauss的看法是， 这对三星来说是首度将基频芯片业务合法化的实际尝试。而对已经感觉到Marvell与联发科(MediaTek)等对手威胁的高通来说，三星新推出的 LTE应用处理器组合芯片又为该公司添了一个烦恼。
“我完全预期高通与其他芯片供货商未来将会被排除在三星的智能手机之 外，”Strauss表示：“不过三星仍有可能会在特定的平板设备采用英特尔等供货商的芯片。”Exynos ModAP的存在甚至在正式发布之前就已经有其影响力，Strauss认为，博通就是因为这款三星组合芯片才决定要放弃蜂窝基频芯片业务。
“我相信博通理解到他们无法打进主流智能手机市场，也肯定无法进驻三星产品；”Strauss 指出：“而且既然三星与苹果(Apple)──也就是高通的调制解调器芯片──主宰4G智能手机市场，对博通来说仅剩非常小的市场，而且大部分是来自新兴市场的手机与平板，这根本不足以支持该公司数百位研发4G技术的工程师。”
本文授权编译自EE Times，版权所有，谢绝转载
本文下一页：三星营收严重依赖手机
相关阅读：
• 中国手机市场分析：本土厂商将唱主角
• 国产芯片将扩大低端智能终端市场份额
• 联发科预热首款4G真8核芯片，详解MT6595技术tBiesmc
{pagination}
Strategy Analytics资深分析师Sravan Kundojjala也同意以上的看法，他指出，除了三星积极进军LTE市场，华为(Huawei)内部的LTE研发活动应该也是促使博通决定放弃基频芯片市场的原因之一：“基带芯片市场非常着重研发。”
Kundojjala 进一步指出：“我们估计博通自2007年以来，已经花了30亿美元进行蜂窝基频技术研发，完全没有利润；而对博通来说，甚至在推出Cat 6/Cat 7/Cat 9/Cat 10 LTE基频芯片与SoC之前就退出市场虽然痛苦，这个决定事后看来是正确的。”
三星营收严重依赖手机
若要让自家LTE调制解调器芯片在市场上占据一席之地，三星必须要扩展自家行动设备产品以外的应用版图。“我们相信三星在LTE基频市场的成功，要看该公司多快 能争取到自家品牌以外的客户，以及能多快精进自家的LTE基频产品蓝图，好赶上高通。”Kundojjala表示：“这需要花费很大的投资与时间。”
能进驻自家品牌产品是很不错，但这样的策略无法长久。三星最近才公布，该公司第二季营业利润滑落至两年来的低点；三星指出这与韩元兑美元与欧元汇率升值有关，但该公司连四季利润下滑的状况也暴露了一个不可否认的事实：三星变得太依赖智能手机。
三星近日也坦承，该公司的中低阶智能手机在中国与部分欧洲市场表现不佳，因为竞争激烈且需求趋缓。无论愿不愿意接受，三星手机部门的衰弱就是会影响到该公司手机芯片业务的表现，当然也包括新发布的LTE应用处理器组合芯片。
本文授权编译自EE Times，版权所有，谢绝转载
编译：Judith Cheng
参考英文原文：Can Samsung Get Qualcomm in LTE Market?，by Junko Yoshida
相关阅读：
• 中国手机市场分析：本土厂商将唱主角
• 国产芯片将扩大低端智能终端市场份额
• 联发科预热首款4G真8核芯片，详解MT6595技术tBiesmc
{pagination}
Can Samsung Get Qualcomm in LTE Market?
Junko Yoshida
Casualty is not Qualcomm but Broadcom
MADISON, Wis. — Samsung took the wraps off a new LTE radio chip this week. The chip, supporting both FDD and TDD, is built on the 28-nm HKMG process. It's noteworthy that Samsung has integrated its own quad-core application processor and called it Exynos ModAP.
In short, Samsung is consolidating its cellular components under the Exynos brand, mimicking Qualcomm's Snapdragon family of mobile chips. The intention is clear: Get Qualcomm.
However, the amount of damage Samsung can wreak in today's totally Qualcomm-dominated LTE market remains unclear.
Samsung's move is destined to design Qualcomm's chips out of Samsung's handsets and tablets. But the LTE-app processor combo chip, which hasn't even hit the market yet, has already delivered the biggest blow. The casualty isn't Qualcomm. It's Broadcom, which announced plans in early June to shut down its cellular baseband business.
No matter how you slice the LTE market, Qualcomm still has a solid lock on the mobile chip market. "Qualcomm still enjoys over 95% of the multimode LTE market, even with new entries announced in 2014," Will Strauss, president of Forward Concepts, told us.
Strategy Analytics reported (subscription required) that Samsung ranked fourth in the LTE baseband market in the first quarter with a revenue share of just 1.7%.
Multi-mode and single-mode FDD-LTE baseband supplier share in 2013.
(Source: Forward Concepts)
Samsung isn't new to the LTE modem business. It has been shipping its own multimode 4G modem used inside its Galaxy 4 and 5 handsets, but the company has done so "only in Korea," according to Forward Concepts.
For handsets and tablets sold in the rest of the world, Samsung has relied on competitors' LTE modem chips. "Depending on the region, Samsung employed Qualcomm 4G modems, while they have employed 4G modems from Intel -- actually, the Galaxy Tab 3 10-inch socket -- and Marvell's and Renesas/Broadcom 4G modems in Galaxy Tab 3 seven-ince sockets," Strauss said.
That picture is about to change.
Bigger impact on Broadcom
The industry views Samsung's announcements about its LTE/Modem combo chip and an LTE-Advanced modem -- both based on Ceva's DSP -- as "a genuine attempt by Samsung to legitimize its baseband offerings for the first time," Strauss said.
To Qualcomm, which is already feeling rivals like Marvell and MediaTek nipping at its heels, Samsung's new LTE-app processor combo chip is one more worry.
"I fully expect that Qualcomm and others will be eliminated from future Samsung smartphones," Forward Concepts' Strauss said. However, "Samsung may still employ Intel and others in select tablets."
The Exynos ModAP's existence was evident even before it was announced. Industry analysts say Samsung's combo chip must have prompted Broadcom to divest its cellular baseband business.
"I believe that Broadcom realized that they couldn't get in a major smartphone socket, and certainly not in a Samsung socket," Strauss said. Further, "since Samsung and Apple -- namely, Qualcomm's modems -- dominate the 4G smartphone market, only small pieces of the pie, mostly the Third World handset and tablet suppliers, were available to Broadcom, a market not big enough to support the hundreds of engineers they have devoted to 4G."
Sravan Kundojjala, a senior analyst at Strategy Analytics, agreed with that assessment. In addition to Samsung's LTE efforts, he said, Huawei's in-house LTE activities must have also weighed on Broadcom's decision to exit the baseband market.
"The baseband market is quite R&D intensive," Kundojjala said. "We estimate Broadcom has spent over $3 billion on cellular baseband-related R&D since 2007 without profit. While it was painful for Broadcom to exit the market before even releasing Cat 6/Cat 7/Cat 9/Cat 10 LTE basebands and SoCs, it was the right decision in retrospect."
Heavily dependent on Samsung Mobile
For its LTE modem chip to become a truly visible factor in the market, Samsung will have to do more than gain design sockets in the its own mobile division's products.
"We believe Samsung's success in LTE basebands depends on how quickly it can diversify its customer base beyond Samsung Mobile and how quickly it can advance its LTE baseband roadmap to match Qualcomm," Kundojjala said. "This will require significant investment and time."
Having built-in, internal design sockets is great. But that strategy can't last forever.
Samsung disclosed this week that its second-quarter operating profit dropped to a two-year low. The company cited the South Korean currency's appreciation against the US dollar and the euro. But the fourth straight quarter of profit decline has exposed an undeniable fact: Samsung has become too dependent on smartphones. It acknowledged this week that sales of its midrange and low-end smartphones were weak in China and some European countries, due to stiff competition and slow demand.
Like it or not, Samsung Mobile's struggle affects Samsung's mobile chip business, including the newly announced LTE-app processor combo chip.
责编：Quentin
扫码分享到好友
