I'va visited Yamar Electronics, (www.yamar.com), I think they had 6 employees at the time, so they are probably the second smallest. Their premises were certainly modest, by any standards. Yamar develop ASICs for DC powerline communication.
Wow, four employees is amazing -- I never imagined such a thing. I'd be curious to see how their business model works. Do they actually manufacture their own ICs? Or do they contract the manufacturing out?
I like how the founder saw that his previous employer was losing opportunities by not looking at this type of work.
I once worked for a privately owned company that was purchased by the mega-corp of danaher. Before that purchase, we did a lot of custom software changes for the customers, including some for older products. The customers loved that we would go out of the way for them, and these changes were lucurative for the company. But, after the takeover, the new mega-corp, who only liked to make optimized, outsourced widgets, decided that we would no longer pursue these custom software opportunities since they didn't fit into their mold.
I still wish that I would have / could have somehow spun off a custom software group to take over that stuff, but that would have required source code access which I'm sure wouldn't have been granted. In the end, the mega-corp couldn't figure out the business and sold the company to one of the original company's former competitors....
Amelco, Fairchild, RCA, TI, and Westinghouse developed early analog ICs in the 1960s. But the Fairchild µA702 op amp, created in 1964 by the team of process engineer Dave Talbert and designer Robert Widlar, was the first widely-used commercial product. Fairchild quickly became the analog leader and over the next 50 years went on to seed hundreds of analog IC companies with their talent, most in Silicon Valley. Today the Valley is teaming with talented, retired and semi-retired analog designers and we have gathered several dozen of the best under our banner to contract to us to develop our custom Analog ASICs.
Unlike other IC companies where the sales guys make the big commission bucks, we incentivize the designers...they get commissions for the life of the product they design. It's only fair, they do the heavy lifting. This leash keeps them tied to the product to continue to improve yields and resolve and issues that may arise in the future.
We outsource wafer fab to large, high quality foundries, mostly located in the US. This minimizes many issues. We have a dozen or so wafer probers here in San Jose and we do 100% wafer sort before sending the wafers off shore to India for assembly and final test and eventually shipment to the customer. We do internal qualification and use external labs when required.
Most test development is done in house, but occasionally, for larger projects (like our new JV700, an ODB-II Diagnostics Interface/Controller, we will contract to have built a specialized tester.
It may sound simple, but trust me, managing the logistics is very tricky.
You hit the nail on the head. When you have investors and shareholders you lose control of your destiny. Everyone expects growth in revenue and profitability (damn the torpedoes, full speed ahead), and while that can be a good thing, it isn't always good for the customer. Look at the Gross Profit margins of some of the leading Analog IC companies....70%, 80% or higher. Selling a product for 4 or 5 times what it costs to manufacture is just wrong. That's not our approach to winning customers. In all fairness, the big Analog guys have to fund invention to stay alive. But let's face it, how many op amps or LDOs does the world really need? Most of what we do is cost reductions for our customers. It's a different mind set.
Thanks for the explanation, BobFrostholm. Even having read your explanation, though, I'm still in awe. As I understand it, you are making your own custom ICs, which means that your people need to do the design work, right? That's still mind-boggling to me, considering how many engineers (I've assumed) are involved in the typical chip design among the big semiconductor companies. Congratulations on being able to make this model work.
Like I said, what makes us efficient is our ability to outsource almost everything. We have about 30 designers that we contract from time to time as the need arises to do the design work and layout. With near zero overhead, our costs are extremely low.
Festo's BionicKangaroo combines pneumatic and electrical drive technology, plus very precise controls and condition monitoring. Like a real kangaroo, the BionicKangaroo robot harvests the kinetic energy of each takeoff and immediately uses it to power the next jump.
Design News and Digi-Key presents: Creating & Testing Your First RTOS Application Using MQX, a crash course that will look at defining a project, selecting a target processor, blocking code, defining tasks, completing code, and debugging.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.