The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: Chuck Simmons
Organization: You jest.
X-Mailer: Mozilla 4.61 [en] (X11; U; Linux 2.0.33 i586)
Subject: Re: Examples Of Common MCU's In Consumer Electronics?
References: <3D7E8B80.firstname.lastname@example.org> <3D80B75F.email@example.com> <3D81375D.918A0030@webaccess.net> <3D879690.firstname.lastname@example.org>
Date: Wed, 18 Sep 2002 00:45:49 GMT
NNTP-Posting-Date: Tue, 17 Sep 2002 17:45:49 PDT
Howard Delman wrote:
> Chuck Simmons wrote:
> > Howard Delman wrote:
> >>Geraldo Sazias wrote:
> >>>"Howard Delman" wrote in message
> >>>>Geraldo Sazias wrote:
> >>>>>Can someone cite some examples of 'common' (frequently used by
> >>>>>MCU's (AVR,PIC and MSP430) being used in consumer and or mass produced
> >>>>>electronics? Whenever I open up some consumer electronics device it has
> >>>>>either an unknown MCU or one I can't identify at all (ussually bump
> >>>>>packaged directly onto the PCB). Is it even economically viable to use
> >>>>>on of
> >>>>>these MCU's in a million plus class consumer electronics device?
> >>>>One reason that you are more familiar with PIC and AVR is that they are
> >>>>actively marketed to hobbyists. The tools are cheap, and the parts can
> >>>>be easily bought in low volumes. Most other manufacturers avoid this
> >>>>market. I used a small 8 bit Mitsubishi MCU a couple of years ago, and
> >>>>the development system was $6000. Not many hobbyists will ever see
> >>>>these parts, but they were right for the application and the $6000 cost
> >>>>was easily amortized over the first year's forecast of 100,000 units.
> >>>>In the past five years, I've designed products with MCUs from Samsung,
> >>>>Toshiba, ST, Philips, and Cypress. Most of these vendors won't even
> >>>>talk to you unless you're looking at tens of thousands of parts a year.
> >>>>Oh... None of these would have their "real" part numbers on them. They
> >>>>would have a part number made up by my client.
> >>>Cool. What kinds of products are we talking about, what prices and what
> >>Toy airplane, coin operated video game, set top box, telephone test set,
> >>network password generator. These parts ranged in price from less than
> >>a dollar (Samsung in toy airplane) to around $5.00 (Cypress MCU w/ USB
> >>in video game). I won't even try pricing a part for volumes less than
> >>1K. The vendors would just ignore me. None of these projects turned
> >>out to be huge, although the Samsung chip (KS57C2308 - 4 bit core, 8K
> >>mask rom, drivers for alphanumeric lcd, 80 pin chip) is likely to hit
> >>100K volume by the end of the year. This is not considered a big order
> >>by Samsung. IIRC, it was quoted around a dollar. The development
> >>system was around $3000. See my web page for more details.
> >>BTW - I have priced PICs and AVRs for new designs, and have always found
> >>them to be more expensive than comparable parts once volumes get high.
> >>They are better in very low volumes. My guess is that they end up in
> >>high volume products because engineers are used to them, and already
> >>have the tools.
> > Not altogether true. I don't know about PIC but AVR is very nice for
> > some ASICs. I'm working on an ASIC now with 3 40MHz mega103 (AVR) cores
> > slightly customized, dual port RAM for AVR program and data memory,
> > analog I/O for the AVRs, a 17X17 shared 25ns MAC with 40 bit accumulator
> > and a gate array for various logic needed for the desired function. It's
> > a pad limited die which is about 60mm^2 (128 pins set the pad ring
> > size). I think volume pricing (100k parts) will be as low as $3 each.
> > The part is not general purpose at all since it assumes some other part
> > in the system can do initial code load to it at power up and it has very
> > few digital I/O pins. It is designed for a specific application and it
> > is doing that application well. The chip is part of a chip set although
> > it could sell alone for some applications requiring up to 6 axis motor
> > control. In its current application, it controls 4 motors and handles
> > miscellaneous chores for the chip that acts as the overall supervisor of
> > the system. AVR was chosen because it was cost effective (small silicon
> > area for what it could do and compatability with the analog process
> > needed to build the I/O system).
> > Chuck
> But that is a totally different situation. The economics of cores are
> nothing like the economics of packaged parts.
However, the price is driven by volume, silicon area and package no
mater what it is. One company I worked for used about 10 million Intel
80196 devices per year. They sometimes bought IBM parts for better
pricing for mature products. The MPU cost was quite low. maybe as high
as $4.00 from Intel and a good bit lower from IBM (IBM was mask
programmed and so quite good for mature products. $12.00 each is the
least I will get off with buying just a few of the lowest end 196 parts
in 68 pin PLCC. That part appears to be silicon area, volume and package
cost. Note that opting for a dip usually pushes the price up. Sometimes
Note that ASICs end up as packaged parts. All of the ASIC parts I see of
the type I described come to me packaged. I usually get 300 or 400 from
a prototype run packaged. Most are thrown away eventually. The last
batch I got is all going into the trash because of a show stopper
regression in the logic. They decapped and repaired two parts for me to
use until a new batch comes out with the fix. The decap and modify took
about 24 hours. I hate handling decaps. Just a touch in the wrong place
and you plow through bonds rendering the part useless. The fix is about
... The times have been,
That, when the brains were out,
the man would die. ... Macbeth
Chuck Simmons email@example.com
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup