Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.misc > #2905 > unrolled thread

Interview with the Old Man of Floating Point

Back to article view | Back to comp.misc

Contents

  Interview with the Old Man of Floating Point RS Wood <rsw@therandymon.com> - 2014-05-20 12:35 +0000

#2905 — Interview with the Old Man of Floating Point

Hacker News linked to this:
http://www.cs.berkeley.edu/~wkahan/ieee754status/754story.html

It's long, but a page of simple HTML with no ads or graphics:
refreshing.

//-- clip
In 1976 Intel began to design a floating-point co-processor for its
i8086/8 and i432 microprocessors. Dr. John Palmer persuaded Intel that
it needed an arithmetic standard to prevent different boxes with
"Intel" on the outside from computing disparate results inside. At
Stanford ten years earlier, Palmer had heard a visiting professor,
William Kahan, analyze commercially significant arithmetics and assess
how much their anomalies inflated the costs of reliable and portable
numerical software. Kahan had also enhanced the numerical prowess of a
successful line of Hewlett-Packard calculators. Palmer, now the manager
of Intel's floating-point effort, recruited Kahan as a consultant to
help design the arithmetic for the i432 ( which died later ) and for
the i8086/8's upcoming i8087 coprocessor.

"Intel had decided they wanted really good arithmetic. I suggested that
DEC VAX's floating-point be copied because it was very good for its
time. But Intel wanted the `best' arithmetic. Palmer told me they
expected to sell vast numbers of these co-processors, so `best' meant
`best for a market much broader than anyone else contemplated' at that
time. He and I put together feasible specifications for that `best'
arithmetic. Subsequently Silicon Valley heard rumors ( not from me )
about the i8087. They were aghast; how could Intel pack all that into a
chip with only several thousand transistors?"

"I have said from time to time, perhaps too cynically, that other
Silicon Valley companies got worried enough to join a committee that
might slow Intel down. It was the committee assembled to produce a
standard for floating-point arithmetic on microprocessors." 
//-- clip

-- 
RS Wood <rsw@therandymon.com>

[toc] | [standalone]

Back to top | Article view | comp.misc

csiph-web