Cost of a microcontroller - is it very important?

A newbie here. After reading a small portion of the discussion, this question comes up i my head, is the cost of microcontroller a very important issue in terms o picking the right one? How much a microcontroller CAN contribute to whole product?

Or is it the functionality, speed, power consumption and developmen toolchain userfriendly-ness more important in the decision?

I am thinking about using Cortex M3 for my server cabinet temp controlle project, is it a bad choice in any context?

Many thanks in advance

Henry

Reply to
henza
Loading thread data ...

Like all things it depends on the end application. I've worked in an Automotive environment and where we were developing a high volme product of maybe 500k units/year for a relatively mature application then a $3 micro. is very sgnificant for a total product selling price of maybe $40. On the other hand I've also worked on a product with a volume of only 1000/yr and product selling price of around $1k so then the $10-$15 micro. isn't so significant. In this case it was a very new technology where factors such as reliability and safety were far more important than cost.

I believe military and medical applicaions tend to be even less cost sensitive for the individual components.

Reply to
Dave

For a server cabinet temp controller project, price doesn't matter. Choose according to which one will teach you the most.

For the kind of things I do (quantities as high as 100,000 units per hour) price is everything. Do the math.

--
misc.business.product-dev: a Usenet newsgroup 
about the Business of Product Development.
     -- Guy Macon
Reply to
Guy Macon

Like Guy said: Do the math.

If you're building 100 of something, _ever_, and you can save $10000 worth of engineering time by spending $20 more per processor, you've just saved $10000 - $2000 = $18000, and you're a hero.

If you're building 100K parts/year, then spending $10000 worth of engineering time to save one thin dime per finished part is a break-even. If you can save a quarter per finished part -- you're a hero.

Now go back to that first case, and ask yourself if spending $10000 of engineering time to save $250 over the life of the product is a good thing or a bad thing...

You should _always_ consider how many parts are going to be made, and how much of a hurry you are in to get working devices on the market. You'll _always_ be faced with having to spend money to save money -- it's just a question of whether the smart money gets spent on a (relatively) huge processor larded with resources to ease the development burden, or whether the smart money gets spent on development to use the thinnest processor possible.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html
Reply to
Tim Wescott

I just did and find them impressive, especially if the manufacturing is 24/7 forever. It must be a challenge to find components in that quantity (0.9 Gparts/year).

I didn't think IC manufacturers could turn out parts at that rate. If they can, I'd think you'd need to be mouth-of-the-factory to keep delivery logistics in hand.

Reply to
Everett M. Greene

up i

terms o

controlle

Thanks guys for the info. Appreciate it!

Reply to
henza

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.