Path: ...!feeds.phibee-telecom.net!2.eu.feeder.erje.net!feeder.erje.net!newsfeed.bofh.team!paganini.bofh.team!not-for-mail
From: antispam@fricas.org (Waldek Hebisch)
Newsgroups: alt.folklore.computers,comp.os.linux.misc
Subject: Re: The joy of FORTRAN
Date: Wed, 26 Feb 2025 14:24:15 -0000 (UTC)
Organization: To protect and to server
Message-ID: <vpn86d$3osdj$2@paganini.bofh.team>
References: <pan$96411$d204da43$cc34bb91$1fe98651@linux.rocks> <lhqvP.1323465$if26.592741@fx13.iad> <vplmop$2lj8$1@gal.iecc.com> <87y0xtjts3.fsf@localhost> <vpm1u1$io3$2@gal.iecc.com>
Injection-Date: Wed, 26 Feb 2025 14:24:15 -0000 (UTC)
Injection-Info: paganini.bofh.team; logging-data="3961267"; posting-host="WwiNTD3IIceGeoS5hCc4+A.user.paganini.bofh.team"; mail-complaints-to="usenet@bofh.team"; posting-account="9dIQLXBM7WM9KzA+yjdR4A";
User-Agent: tin/2.6.2-20221225 ("Pittyvaich") (Linux/6.1.0-9-amd64 (x86_64))
X-Notice: Filtered by postfilter v. 0.9.3
Bytes: 2922
Lines: 34

In alt.folklore.computers John Levine <johnl@taugh.com> wrote:
> According to Lynn Wheeler  <lynn@garlic.com>:
>>John Levine <johnl@taugh.com> writes:
>>> The VAX was developed over a decade later, when they put thousands of
>>> transistors on each logic chip and thousands of bits in each memory chip. It
>>> suffered from a severe case of second system syndrome, where they started from
>>> the elegant PDP-11 and added every feature a programmer could ever possibly
>>> want, with less than fabulous performance to match. There's a reason that the
>>> VAX inspired RISC systems.
>>
>>I've claimed that John Cocke did RISC/801
>>https://en.wikipedia.org/wiki/John_Cocke_(computer_scientist)
>>https://www.ibm.com/history/john-cocke
>>The effort to develop RISC began in 1974, when IBM tasked Cocke and a
>>team of researchers with creating an exchange controller to automate
>>telephone switching -- ...
> 
> The Berkeley RISC work started in 1978 but as far as I know the 801 and RISC
> projects were unaware of each other. They made similar observations that most
> programs used only a small fraction of the instructions on the computers they
> were using (Vax and 360 resepectively) so how about building something simple
> that just had the useful instructions and ran really fast.

It is interesing how different people came to different conclusions.
I have a book by Andrew Tannenbaum where he pitches "high level
machines" implementing complex instructions via microcode.  He
also cites research showing that low end IBM machines can perform
many times faster when compiler generates internal microcode
instead of 360 instructions.  But he did not notice that those
microcode experiments suggest building machines which can execute
simple instructions quickly (that is RISC) and not the "high
level machines" that he pitched.

-- 
                              Waldek Hebisch