NEW YORK Ė How many people does it take to run a mobile chip company these days? Anyone care to guess?
Well, Iíve checked, and apparently you need 4,000 to 5,000 employees.
Until recently, I had no idea on how big an engineering team a mobile chip vendor needs to develop a mobile platform that covers both "global" modem technologies and apps processors.
I was stunned when I heard a Renesas Mobile executive rattling off the number of people he said a mobile chip company needs. Specifically, he cited the manpower needed for everything from the actual development of a modem chip to pushing the modem chip out in the field, to getting an operatorís certification.
I followed up by talking to more people in the mobile IC industry, each time popping the "how-many-people-does-it-take-to-screw-in-a-light-bulb" question. They all confirmed that 1,000 engineers arenít enough, especially if youíre gunning for the global market.
In early 2009, when ST-Ericsson was born as a result of the merger of Ericsson Mobile Platforms and ST-NXP Wireless, the announcement said 8,500 people would be working for ST-Ericsson.
In April, struggling ST-Ericsson said it would cut 1,700 jobs, or 25 per cent of its workforce, including some senior executives, to reduce administrative expenses. As drastic as this sounded, it would still leave ST-Ericsson today with close to 5,000 employees. After STís announcement on Monday (Dec. 10) that it will exit the ST-Ericsson joint venture in the third quarter of 2013, observers in the financial community insisted that ST-Ericsson, if itís serious about finding a buyer, still needs to shed more workers.
Renesas Mobile, similarly distressed, also employs close to 4,000 people.
MediaTek, 65 percent to 70 percent of whose revenue is generated by the mobile business, employs about 6,600 people. In other words, MediaTek, has more than 4,000 employees committed to mobile technology development.
Maybe not so much lack of engineering skills, but maybe thinking in Europe needs to change. Clear out the banking type scandals, have smaller companies with less red tape, get the lazy off social life support/ welfare, and actually contribute (ie. Don't outsource to China at costs that don't allow them to produce quality, and then sell the stuff at German quality prices). Markups are too high in Europe or the UK (and Oz), so they will never be able to compete with Asian suppliers. Tax rates and the dole don't help either -- its that simple.
Ideally, you would expect a product company to have solid engineering/management skills, but in most cases they have minimal skills and talent (focus more on profit margin and staying afloat and Wall Street expectations, no more focus on real enegineering talent), and so there are plenty of repeated mistakes along the way.
That was my original thinking...but then, when you think about all the other telecom operator-related stuff you need to do (beyond chip design)especially when the world has so many different cellular standards with different flavors, it suddenly dawned on me that this may be a business that NO chip company wants to be in. Seriously, it appears to require a lot of people to address the "global" cellular market.
ST-Ericsson have never seemed a solid company to me. They seemed a mish-mash of several divisions that were formed by necessity rather than some unique ideology such as what Apple have.
ST should be even more ruthless and define a very clear ideology of how they will re-dominate.
The point of the story is this: As more and more chip companies are asked to do everything from chip design to software development and designing a "platform," chip companies ought to get paid more or find another bis model ( or simply going out of business.) Especially in a highly competitive mobile chip world, this sure ain't looking sustainable.
No doubt ST's exit is a major blow to Ericsson. It would be very interesting to see who would want to partner with them now to fill ST's void. Or who knows, Ericsson will shut the handset modem shop altogether.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.