Qualcomm Looks To Gobble Up NXP

Remember when we talked about NXP merging with Freescale to move into the top ten semiconductor companies? Yeah, that was just eighteen months ago and just barely closed before the new year. Now it looks like Qualcomm wants to acquire NXP to the tune of $30 billion.

You’re most likely familiar with Qualcomm as a cellphone silicon company. The acquisition of NXP opens up a lot of additional markets with their portfolio of chips — automotive among them thanks to the Freescale merger. Now you should be asking yourself just how big Qualcomm is already. What’s perhaps most interesting is that, as mostly a wireless chip company, Qualcomm is ranked number three in worldwide semiconductor sales. Adding NXP — a behemoth now in the top ten — adds at least 30% to Qualcomm’s numbers.

And so here we are, one step close to a monolithic chip fab that produces all computing power for the human race. Yippie!

31 thoughts on “Qualcomm Looks To Gobble Up NXP

    1. Good idea, but for the 410e it’s basically only a PR and a web page:
      https://www.qualcomm.com/products/snapdragon/processors/410e
      The “Detailed Specifications” is only a product brief without additional information. Not documentation, no development board, no stock information, no price information, no link to distributor product page.

      I would love to uses this Qualcomm Snapdradon 410e or the Samsung Exynos 7570 on the designs I propose to my customers, but it’s actually too hard to get the parts with plausible price and a reliable availability. Even working with Allwinner (as I do now) is more easy, with lot of dev kits with schematics (Orange PI line), at least a bar minimum documentations on all aspects, low price (~4$ for the quad cores H3), and shipping directly from the manufacturer even for a single MOQ. Allwinner support and community is too little but it’s the only bad point.

      The 410e would save the ESP8089/ESP8266 for the WiFi and the FPGA for the dual cameras MIPI CSI to DVP, but it miss the Ethernet MAC and integrated PHY of the H3. OpenCL support is a good potential for video image processing. Not clear if the 410e support OTG. High pin count BGA with 0.4mm ball spacing require high level PCB classes and a lot of time to be routed properly, even if the pins mapping on the BGA is done in a very cleaver way as in the Allwinner H3.

      1. Finally found this afer a lucky google on “760NSP” and “APQ 8016”. Hell, why using a so obfuscated reference for a part called “410e” ?
        https://developer.qualcomm.com/qfile/28813/lm80-p0436-7_b_apq8016_devicespec.pdf
        https://www.arrow.com/en/products/apq-8016e-1-760nsp-tr-010/qualcomm
        It that too hard for Qualcomm to provides those links on there product web page ?
        Price is ~3X the H3 ! Even with the additional external WiFi and FPGA, the H3 don’t cost so much.
        I fail to find a good cheap open source dev board to test and hack the product.

        1. Well, reading the specifications, the 410e:
          * Don’t have integrated WiFI/Bluetooth nor integrated GPS/Galileo/GLONASS. It require additional external chips that increases the price and complexity of the PCB.
          * Have 2 MIPI CSI with respectively 4 lanes (13MP) and 2 lanes (8MP), good point.
          * Might require a complex power management chip like the PM8916.
          * No integrated audio DAC,
          * Integrate a extra Cortex M3 core for the power manager. curious feature.
          * No USB OTG.
          * Do not use SWD instead of JTAG,
          * The 0.4mm ball spacing is in fact a 0.5657mm 45° array of implemented balls at the max. Good compromise.
          * The pin assignments on the specification only show the a single quadrant of the chip ! Have to guess 3/4 of the pins from a lot of tables ? Oh! there is finally a useful link https://developer.qualcomm.com/hardware/dragonboard-410c/tools

          I now wonder if 410e equal 410c… I have evaluated the 96Boards 410c some months ago but I found it so unstable and without usable support that I have give up after having to sort my screens to find one that work with his blurry HDMI/DVI signals. The overall BOM for a working components set is too high.

          I don’t think it’s useful for me to continue to invest time is this kind of solution.

  1. I was really hoping for a more detailed analysis on this acquisition than “QCOM was big and now it’s bigger and big is bad”. Why can’t we have an open hardware community that isn’t so willfully ignorant and spiteful of the industry?

    1. Because qualcomm is well known to have most/all datasheets behind NDA and wont even talk to you at the purchase amounts of hobbyists and small companies. 100k might be too little.
      Even NXP buying Freescale was bad for hobbyists, as NXP put the previously free devtools behind a paywall (IIRC 1600usd, yay).

      Only thing I’d buy from NXP is RF transistors and regulators.
      And even then I’d look at other options first.
      Like Infineon, Qorvo, M/A-Com and Skyworks for RF. TI, Linear and Microchip for regulators.

      Sad to see things going from bad to worse.
      At least NXP has public datasheets for some of their components and it’s not as bad as Beken or RDA micro.

    1. Nah, just a 1000 degree hot end that you can feed sand and a 7 nanometer print resolution would do it…. :-D

      Though if you’re really, really patient, you could use 3Dprinter code/control and pick and place software, and a homebrew STM and build yourself an 8080 over the course of a few months.

  2. I for one welcome our silicone overlords….. All joking aside, large companies will always try to swallow one more piece of the pie. The problem is that once they become bloated they can no longer function. The larger a company gets the more efficient it gets, to a certain point (even that is debatable). Then you have a sharp decline in efficiency, jaded employees and people who hate them. At some point they choke and become loosing proposition. All driven by board of directors from the top which will promptly do something to gut the company and walk away with 90% of its worth leaving it broke and expired. This is the trend in every industry. Get used to it. I have worked in plenty of places that went through mergers. None of it is fun, and it is a sure way to loose talent leaving behind a mess and high expectations that are never met. I have worked in places where the bureaucracy was so large that it took months to get anything remotely simple done. This lead to the place falling from 1st place in their market to being acquired by a competitor that 5-6 years ago was nearly dead last as everyone in the market wanted nothing to do with the mess. The board of directors made $$$ with 10 or more zeros behind it and the company lost 30% of its market share in less than 9months, My advice is don’t invest in either company.

    1. I fear that you are right. :-/ I worked (as a trainee) in a firm (in Europe) that is part of a *really* big compagny which headquarters (and big bosses) are in the US. I needed an update for my computer. Instead of simply give a call to the IT-guy on-site or go and see him personally to talk about the thing i had to write an email to some country in Asia iirc where the request was processed and an order was given to the IT-guy. WTF?!?! I didn’t need 1000000$ of investment, just some more RAM to make a bloatware run… Another day they needed some component that was obsolete and no longer sold by the big dealers or the manufacturer of the machine. By chance i was aware of a dealer selling this thing (value like 10$) but the engineer told me that they can’t buy there because the shop is not already in the system and it would require to much paperwork. I don’t remember exactly how this finished, i think they bought it elsewhere and payed much more (like 100$)???? Well…
      Another problem is that the guys at the top of these massive compagnies are managers or something like this and don’t know anything about the technology. They always want more and more $$$ but there is a limit, machines can’t run 100% of the time (maintenance) and somewhere quality will decrease (and maybe higly specialised employees will go away because there is to much stress).
      So yeah, to big is not good, neither for employees nor for customers. Some competition is always good.

  3. Disclaimer: I’m an Engineer at Qualcomm, but I absolutely do not speak for the company.

    I agree that there is diminishing return on the size / efficiency relationship. I have definitely experienced issues with IT, HR, and other support services lacking efficiency, and what I would considered poor decision making when it comes to offshoring support. Times are tough all around, buck up buttercup; you get over the BS, because it’s the price of working inside in the air conditioning.

    I take some issue with Bruno’s statements though. While I don’t always agree with our senior management’s decision making (I mean I keep asking for a 100% raise, and they keep saying no), I do commend them on their handling of our recent dealings with Janus (the investment group). It appeared to most of us that it was Janus’s desire to get enough leverage to chop us up, harvest IP, and sell off the rest before dumping their shares and leaving us engineers scrambling for something else. The top level folks did what they had to do to be responsive to investors (their bosses), while keeping us mostly whole. Many people were affected, mostly negatively, but we weren’t laid bare and picked clean.

    On the topic of purchasing NXP, I see where it makes sense. There is almost no overlap in our current markets and this enables us to service those markets. We aren’t perfect, but we make a damn good product all in all. Also, my stock went up, so I don’t have to eat the kids this winter.

  4. aaag, whats up with all these acquisitions? I haven’t seen any lately that have made better components or better for the customer in general. Every time someone gets swallowed, you wonder about needing to design out that component for fear of that part dropping out or getting put into a 22+ week lead time so it might as well be dropped. Only the lawyers and financial guys putting the “deal” together seem to benefit, leaving a shell of where a decent company used to live.

    1. Agreed nothing good for the consumer usually comes from these mergers and when you finally end up with a just handful of companies in a market you get stagnation and inflated prices.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.