Signals & Noise: March 28, 1996
to send a letter to signals & noise via email: |
In "Sorry, we’re out of stock" (EDN,Nov 9, 1995, pg 9), Steve Leibson mentions supply changes in electronics parts. I have been in the business for nearly 20 years, and I can’t recall a period when something was not hard to obtain. However, the problems facing the industry today are different.
First, around 80 to 90% of the world semiconductor output is sold to direct accounts; only the residue goes via distribution. In the wider profit-oriented picture, distributors hardly matter (much as the distributors would deny this). When there is any shortage, distributors get only what is left. This situation has gotten worse over the past three years.
In the past 10 to 15 years, there has been a definite shift away from multiple sourcing. Many new devices used to be second-sourced from day one. Now, nearly every new device in the analog field and every new digital product (except a commodity-memory part) are proprietary. Manufacturers make a greater profit this way. These two problems provide a recipe for disaster unless, as Leibson suggests, designers go to great trouble to use multisourced parts.
P Holy
KK Systems Ltd
Shoreham by Sea, Sussex, UK
"Sorry, we’re out of stock," (EDN, Nov 9, 1995, pg 9) stirred up quite a reaction in this very intense organization. Our president, Bob Swanson, took great exception to Leibson’s advice to "use proprietary component features as sparingly as you can," because our company has executed a program to support all users of our proprietary devices. This program includes small companies that rely on distributors.
Our competitors did the opposite, cutting off smaller users and putting all resources into supporting larger companies. Our company has 10,000 customers that are served through distributors, and we have not failed any of them, even though we turn over 15,000 line items of analog products each month.
Roy Twitty
Linear Technology Corp
I read "Sorry, we’re out of stock" (EDN, Nov 9, 1995, pg 9) with more than a passing interest. Like many semiconductor suppliers, Philips Semiconductors Microcontroller Product Group was caught with demand significantly exceeding supply. People assume that these are "good times," but it has not been fun dealing with allocation issues. We believe our increased on-line capacity will restore balance next year.
We have been consistently prioritizing proprietary products over multisourced products. We want to serve customers who design proprietary micros first, because those customers have no other place to go. Contrary to the advice in your editorial, our proprietary micro customers are better off in periods of short supply with our products.
David Fair
Philips Semiconductors
dfair@scs.philips.com
We have received a number of calls in reference to "Windows-based EDA tools: shifting into high gear" by Jim Lipman (EDN,Feb 1, 1996, pg 42) asking why we were not listed among the other EDA companies with Windows products.
For your information, we have signed Xilinx as an OEM partner, and our software will be included on each of its CD-ROMs. Also, based on existing and in-process OEM deals, we will ship more than 5000 Active-CAD Windows-based systems in 1996. Finally, after a thorough evaluation of EDA Windows products, one of the top three EDA vendors has made us an acquisition offer. Things like that do not happen to a nonexistent company.
Stanley M Hyduke
Aldec
Newbury Park, CA
"CCD cameras and frame grabbers" (EDN, Jan 18, 1996, pg 73) cites a price of $125 for a Hitachi VK-C25 CCD camera. The VK-C25 costs $125 (50,000). The module is also available as a finished camera with a case, a microphone, RS-232C or local control, and a parallel digital-video interface. Camera price is determined on case-by-case basis. Contact Hitachi America Ltd at (800) 285-1601, ext 05.
The phone number for Eastman Kodak in "CCD cameras and frame grabbers" (EDN, Jan 18, 1996, pg 73) should be (800) 462-4307.
Ian Poynton and Ron Lorenz’s Innovator write-up in "EDN’s 1995 Innovator and Innovation Competition," (EDN,Feb 1, 1996, pg 79) contains an error. "The absence of overlapping conduction in the output" should read "the presence of overlapping conduction."
| Sound off |