If Intel, still one of the most capable chip firms in history, couldn't produce 5G chips despite spending billions, the standard is clearly too complex.  Someone needs to step forward and reign in 3GPP, brilliant engineers who clearly are unable to serve consumer interest. In practice, GPP has been throwing in far too many requirements in order to win consensus. 

My best guess is the complexity adds perhaps 5% to the cost of 5G in the first few years. If others matched Qualcomm's royalty demands, that would add US$20-30 to the cost of the phone. Take your estimate of revenue, do the arithmetic, and realize how big the issue is.

Intel was a year behind Qualcomm in 5G. Beating Intel required some of the most intense engineering in history.

It spent as much as US$4 billion per year, but after the Apple-Qualcomm deal have now given up. Strategy Analytics, a reputable firm, estimates Intel directly lost US$14B. I add the reputational loss for my US$20 billion figure. 

Qualcomm now has a dominant position in the West and a strong patent position everywhere. The Qualcomm tax is now set at a higher rate, 5% of the total price.  Qualcomm deserves a strong return but not an unreasonable royalty.

 Intel Gives Up on 5G Modems While Qualcomm Scores Big with Apple

On April 16, 2019, Apple and Qualcomm settled their bitter legal battle just one day after opening arguments were exchanged in a US Federal court. On a momentous day, Intel also announced its intention to exit the smartphone cellular modem market.

  • There are some clear parallels between the Nokia-Qualcomm and the Apple-Qualcomm lawsuits. The Nokia - Qualcomm lawsuits lasted for three years before the settlement in 2008 (Apple - Qualcomm lasted for two years). Both Nokia and Apple were the smartphone market leaders when these lawsuits were initiated.
  • When Apple and Qualcomm announced their settlement, Apple’s key modem supplier Intel announced its exit.  Just like when Nokia and Qualcomm settled in 2008, TI, Nokia’s key modem supplier at that time, announced its exit from the market.

Intel’s cellular modem journey has been rocky; Intel sold its XSCALE mobile processor product line to Marvell in 2006, and then re-entered the market in 2010 with the acquisition of Infineon’s cellular chipset business. In addition, Intel was unable to expand beyond Apple in the past three years. Qualcomm’s settlement with Apple means Intel would have to share iPhone volumes with Qualcomm and this would have put further pressure on Intel’s financials. With a single customer and lower volumes, Intel would not have been able to justify sustaining its 5G investments.

Intel’s LTE baseband shipments reached an all-time-high in CY 2018, thanks to iPhone design-wins. Strategy Analytics estimates that at this peak Intel held just 8% of the global baseband processor CY 2018 market.  The baseband market grew 1 percent year-on-year to reach $21.4 billion in revenue in CY 2018. Qualcomm (49%), MediaTek (14%), HiSilicon (13%), Samsung LSI (13%) and Intel (8%) were the top-five baseband players, in terms of revenue.

  • At its peak, Intel spent over $4 billion on mobile R&D annually. However, since 2015, Intel’s mobile R&D run-rate came down as the management prioritised LTE slim modem development and cut down on the SoFIA integrated apps processor project.
  • We estimate that Intel has lost ~$16 billion on mobile between 2011 and 2018. In contrast, the market leader Qualcomm makes up to 20 percent operating margin in its wireless chipset business.
  • In recent years, Intel has shifted its focus to data-centric business and the company is unlikely to see much impact from its modem exit. If anything, Intel could benefit from such a move as it reduces its R&D burden, giving the company an opportunity to focus more on the 5G infrastructure side of things. Intel could potentially sell its modem IP to Apple as both companies already have a good working relationship.

Qualcomm is likely to be the sole modem supplier in the new iPhones to be announced in September 2019. This exclusivity will continue unless and until Apple comes up with a 5G baseband design of its own. We assume that Apple will be able to accelerate its 5G iPhone design with the help of Qualcomm’s 7 nm-based X55 multi-mode 5G modem. It is quite possible that the 2020 iPhone will feature a 5G variant of Qualcomm’s X55 modem. We believe it is unlikely that Apple will release a 5G iPhone in 2019 as the company would have to opt for an inefficient design (dual basebands (X50 5G single-mode plus X20 4G multi-mode) and an A13 apps processor), which could affect the device costs, form factor as well as power consumption. We believe iPhone 5G is a 2020 story and will feature Qualcomm multi-mode X55 5G modem as well as Apple’s A14 apps processor.

With this approach Apple can launch a 5G phone and then design its own 5G chipset if it wishes. When ready with its own 5G chipset, Apple could drop Qualcomm’s modems in the future. We note that Apple has insisted on having multiple modem suppliers, and it is likely that the company will explore alternative options to Qualcomm if an opportunity is presented. Apple could potentially acquire Intel’s 4G/5G modem IP assets and talent to boost its modem design credentials and to lessen its dependence on Qualcomm in future.

From Qualcomm’s point of view, the settlement with Apple is a huge victory. Qualcomm went through a turbulent period in the last two years in dealing with Broadcom’s attempted hostile takeover, the failed acquisition of NXP, the Apple legal battle, a legal fight with Huawei, regulatory lawsuits (TFTC, JFTC, KFTC, EU and FTC), and internal cost cutting and restructuring. We believe Qualcomm’s increased semiconductor content (antenna to modem) in the first 5G iPhone could drive strong incremental revenue growth. We expect Qualcomm to scoop up the market share left by Intel. From a baseband and chipset competition point of view, Intel’s move further strengthens Qualcomm’s market share and dominance. Qualcomm also stands to gain over $6 billion as one-time payment, which includes catch-up royalty payments for the last two years as well as 5G licensing fee. In addition, Qualcomm’s product and royalty revenue from Apple will see a big boost in FY 2020 (starting in September 2019).

dave ask

Newsfeed

Samsung has delivered 5G chip samples to BBK's Oppo and Vivo, the #2 or #3 phones manufacturer. Samsung is facing off against Qualcomm in the 5G market. Qualcomm is unfazed and reportedly moving the production of their next chip from TSMC to Samsung. 

Sprint's 2.5 GHz 5G is delivering 100-500 megabit downloads consistently. That bodes well for China Mobile, using the same frequencies.

Vodafone, BT, and soon 3UK are delivering modestly sized 5G mid-band networks. Vodafone is also live in Spain and Italy. 

Sunrise in Switzerland is using 5G mid-band for fixed wireless in rural areas.  

More newsfeed

----------

Welcome  1,000,000 Koreans bought 5G in the first ten weeks. The demand is there, and most of the technology works. Meanwhile, the hype is unreal. Time for reporting closer to the truth.

The estimates you hear about 5G costs are wildly exaggerated. Verizon is building the most advanced wireless network while reducing capex. Deutsche Telekom and Orange/France Telecom also confirm they won't raise capex.

Massive MIMO in either 4G or "5G" can increase capacity 3X to 7X, including putting 2.3 GHz to 4.2 GHz to use. Carrier Aggregation, 256 QAM, and other tools double and triple that. Verizon sees cost/bit dropping 40% per year.

Cisco & others see traffic growth slowing to 30%/year or less.  I infer overcapacity almost everywhere.  

Believe it or not, 80+% of 5G (mid-band) for several years will be slower than good 4G, which is more developed.