I asked a friend that question a few months ago and he said the design cycle to get into a mobile phone is about a year. But once you are part of a design platform, then you can get into other products quicker.
You would be surprised as to how much this costs. Even Intel will feel the pain and they can fail. This is very specialized knowledge that few companies can buy or develop. There are a couple of companies that spent over $1 billion and failed. Qualcomm spent far more than that.
I think several of those are going to bite the dust too. Some of these companies have a product but it will not be enough to compete in the market. Broadcom was always hyping what they had but they knew they had nothing. Some other companies are doing the same thing. It could come down to only three or four vendors. I wonder what Intel has, or Spreadtrum, or Marvell, or Mediatek. Erricsson claims they have a good part. With all the bs being said, it is no wonder nobody knows what to think about the market. Fewer vendors will give better pricing power to the few companies that can compete. When the market starts to reduce the numbers of wireless standards the vendors have to support, then things will get easier. I have reasons to think that Mediatek, Marvell, Intel and Spreadtrum will all fail. I am waiting to see if that happens.
The original acquisition took place quite late i.e. after Renesas Mobile declared closure of the business. Nothing happened while it was set for sale. This clearly indicated that future prospects of the platform was good. And this is with good reason as Broadcom was getting a market ready modem which they desperately needed.
What I think happened next was that the projects did not take off as promised by the potential phone maker. If you have worked in this industry, you can guess who it could be.
That was also one of the reasons why Renesas Mobile decided to quit after Nokia went for Qualcomm/Windows and the promised volumes didnt ship.
This is indeed quite shocking news... I cannot explain how come B'com has now 3000 enginners behind this chipset, when they bought from Renesas about 1500 engineers, hence double headcount after all the layoffs on both sides. If this is the situation, it might make sense that the economics behind is not adding up. Then indeed, it is not easy to make miraculous changes in this business in 8 months! I wonder when the mistake has been made, when B'com bought Renesas?, or now when they pulled out of this business?
But what is clear is that in less than 1 yr, two players are out, Renesas and B'com. Who's next? Intel? Nvidia? On the prospect that somebody can buy B'com's chipset... hard to believe. There were comments on Apple, the main question in this case is: if Apple did not buy the B'com product, why would they buy the business itself? I'am afraid this is the end of the line, RIP NWM!
I definitely agree with your point. Original acquisition doesn't seem well thought out now; They would have been really wrong in their calculations if they expected to start making dent into the modem space within 8 months.
They will inherit multiple modem platforms of the same generation from Broadcom and Renesas Mobile just like it happend with STE. If they were serious about vertical integration, they would have bought Renesas Mobile last year or someone else by now. Its too messy at Broadcom.
A Book For All Reasons Bernard Cole1 Comment Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...