Yes it is quite practical to have chips and evaluation board's cost below $100 so that engineers can buy it themselves instead of waiting for long process in corporates to get the funds approved. Have seen the frustration of engineers due to delay in funds approval.
Long before the Maker movement became popular, I was a PR manager at Signetics. Many engineers and tech editors happily recounted stories of receiving free 555 timer samples from the company for use in school projects or garage experiments. These giveaways helped make the 555 one of most popular devices of all time.
Even though, Signetics was not an epic success like Intel, it did help educate generations of engineers and find millions of homes for the venerable timer. The moral of the 555 story is that looking at the immediate bottom line is not the only way to evaluate the love fest between makers and chipsters.
I wonder if today, instead of making all the money in 1M units/year chips ,some ic companies make a larger part of their money in 50-100kunits/year sales, since more things are sold at that levels and at those levels companies care more about time to market and less about every cent. If this is true and growing,it's another reason for the maker strategy to make sense.
@susan.rambo: That's a good way to describe it. Also getting product out to makers is like advertising/marketing -- hopefully one of those makers will be an influential engineer
Engineers tend to use parts with which they are familiar. If a prototype is developed on particular hardware, that hardware will be the default for going to production. As long as the chip vendors can minimize the support costs for a lot of startups by handing them off to groups of enthusiasts it's hard to see this as a worse gamble than creating reference designs.
If nothing else it creates a way for them to get deeper into retail channels. These maker components are showing up on the shelves at Fry's, which makes them much more accessible than having to go through the distributers.
When I talked to a Marvell engineer at EE Live! last month, I asked him what his company is planning to do between now (they were about to close the Indiegogo thing) and September (when they actually start shipping Kinoma Create).
His answer was:
"we hope to ensure enough documentation, sample projects, and example applications for Kinoma Create users to dive into."
Based on your comment, apparently, that's exactly what chip vendors must do.
IMO the key to success in selling lots of chips to lots of customers is excellent documentation. Some companies have the philosophy that "documentation is expensive" and write terrible, insufficient documentation, so even if the technology is excellent underneath you'll need so much tech support to use it that the vendor can only support a few large customers. Meanwhile, your competitor can put excellent documentation out on the Internet, for free, and as if by magic the they will sell oodles of chips -- a few at a time to each customer, at least initially -- but those individual chips add up. JMO/YMMV
As someone who loves to work and play with technology, my chief frustration is insufficient documentation so I am unable to do what I want with the chips. There are far more parts out there than I or any engineer can possibly evaluate, so if I start running into roadblocks it's often much easier to switch to a different part than to try to overcome the roadblocks. If there's sufficient open documentation, I can figure out what's going on myself and share that knowledge with others at the appropriate forum.
@joe.raffa. Thanks for your comment. Yes, "spray and pray" seems to be exactly where this seems to be going. The question is then how best to cast a wider net.
I heard from a chip executive I was interviewing in Beijing last month...as soon as he posted a reference design (of his new chip for wearables) on WeChat (China's widely popular messaging board), he received a ton of request for people wanting that board.
It tells us that 'makers' -- no matter where they are located -- are hungry and the social media definitely helps.
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 ...