Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!rutgers!sri-spam!ames!amdcad!phil
From: phil@amdcad.AMD.COM (Phil Ngai)
Newsgroups: comp.sys.ibm.pc
Subject: Re: Intel Microprocessors (History)
Message-ID: <17979@amdcad.AMD.COM>
Date: Thu, 20-Aug-87 12:55:53 EDT
Article-I.D.: amdcad.17979
Posted: Thu Aug 20 12:55:53 1987
Date-Received: Sat, 22-Aug-87 10:45:59 EDT
References: <1112@lznv.ATT.COM> <399@aucs.UUCP> <3225@cucca.columbia.edu> <892@looking.UUCP> <79@LBI.UUCP> <182@hobbes.UUCP>
Reply-To: phil@amdcad.UUCP (Phil Ngai)
Organization: Advanced Micro Devices, Inc., Sunnyvale, Ca.
Lines: 45

In article <182@hobbes.UUCP> root@hobbes.UUCP (John Plocher) writes:
>Hindsight is always best.  Granted, segment registers can be a pain, but
>AT THE TIME, they were very useful!  What were *you* doing with micros in
>September of 1980 (when the PC was starting design life)?

John does a decent job of defending IBM's choice of Intel's 8088 for
the PC. I'd like to add a few notes on the environment in which Intel
designed the 8088. (Of course, I hate small segments as much as anyone
but I suspect a lot of the youngsters on the net don't have enough
sense of history to understand why things are the way they are.)

These quotes are from William Davidow's book _Marketing High
Technology_.  He was senior vice president of sales and marketing for
Intel and has a Ph.D from Stanford. 

 "Les Vadasz and I had been co-general managers of the microprocessor
division in 1976 when the 8086 was being planned. At the time we
decided to make the product an extension of the then-successful 8080
family. That created some design problems, but they were more than
counterbalanced, in our opinion, by the resulting access to a large
existing software library. 
  The 8086 was introduced to the market in 1978. As the first
high-performance, fully supported 16-bit microprocessor, it had
quickly gained the top position in the market, capturing the lead from
older and less capable products supplied by TI and National."

I did a lab project in 1978 using the then new and expensive 16,384
bit dynamic RAM. I needed an 8-bit memory but the RAMs were so
expensive I could only afford one and I had to multiplex and
demultiplex 8 bits into 1 bit. From 1980, when I started working,
until about 1983, 65,536 bit dynamic RAMs were state of the art. 

Our era of cheap megabyte memories started a very short time ago.  The
8086 was designed over 11 years ago, in a time when 4,096 bytes of
static memory took up an entire S-100 card, and dynamic memory was
regarded as unreliable. The designers also had not had our experience
of more than a decade of watching memory sizes double every few years
and offering a megabyte of address space must have seemed like quite a
bold idea. 

Hey Clif, why don't you defend your own company?
-- 
I speak for myself, not the company.

Phil Ngai, {ucbvax,decwrl,allegra}!amdcad!phil or amdcad!phil@decwrl.dec.com