Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10 5/3/83; site t4test.UUCP
Path: utzoo!watmath!clyde!floyd!harpo!ihnp4!zehntel!hplabs!intelca!t4test!murray
From: murray@t4test.UUCP (Murray Lane)
Newsgroups: net.micro
Subject: Re: chip availability
Message-ID: <492@t4test.UUCP>
Date: Tue, 10-Apr-84 16:43:58 EST
Article-I.D.: t4test.492
Posted: Tue Apr 10 16:43:58 1984
Date-Received: Thu, 12-Apr-84 05:17:20 EST
References: <423@dual.UUCP>
Organization: Intel, Santa Clara, Ca.
Lines: 56

There has been a recent spat of articles dealing with chip availability. 
I've decided I can't ignore them any longer. Here are some VERY unofficial
answers to some of the questions. They do not represent any official
stance by Intel or any other chip manufacturer. As evidence that I know
what I am talking about, a portion of my job is to get 80286 family
produced in quantity.

>	Topic for discussion: WHY is the supply of chips so short right now?
>	If people were making chips at the rate they were a few years ago,
>	there would be plenty of parts for everyone.
>	

Semiconductor manufacturers are making chips at many times the rate they were
a few years ago, the problem is, demand has gone up by an order of magnitude
more than production capacity.

>	My theory: most of the chip mfgs got caught with their pants down in the
>	latest recession. They were building along at a merry pace and all of a
>	sudden, BANG, nobody was willing to carry inventory any more. Where most
>	people carried 6 months or more parts stock, all of a sudden they were
>	trying to pare down to only keeping this weeks' production requirements
>	in stock. As a result, the chip mfgs. got left with a huge oversupply.

I cannot comment about the rest of the industry, but at Intel we DID get stuck
with a huge inventory in the 1974 recession. Intel did not get burned in the 
last recession, management saw it coming and pared down the inventory. I don't
know about our customers inventory.

>	They survived this, though a look at financial reports from a few years
>	back will show that it wasn't easy on their pocketbooks. It seems that
>	this time around, they are going to throttle the supply so that they
>	won't get burned the same way if another recession hits - they will
>	be building at somewhat under the demand so that if a slump hits, they
>	will be building right at or a just a little over the demand. Besides,
>	asupply which is less than the demand would tend to drive the prices
>	up, right?

At Intel, on the 80186 and 80286 families, we are currently producing at
several times what we were projected to be doing at this time a year ago.
However, demand has shot up at an even greater rate (actual numbers
are propriatary). We are ramping up production as fast as we can (we are
on allocation from some of our vendors too). Could we have built up sooner?
Not likely, Intel has been building wafer fabrication facilities all through
the recession, but we are still short of wafers.

>	As I say, only my theory, but it makes sense to me. Any comments?
>	
>		    Mats Wichmann
>		    Dual Systems Corp.
>		    ...{ucbvax,amd70,ihnp4,cbosgd,decwrl,fortune}!dual!mats

Sorry, as one deeply involved in trying to meet customer demand, your theories
so not make sense to me.

				Murray at Intel @ t4test