Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site aurora.UUCP
Path: utzoo!watmath!clyde!cbosgd!cbdkc1!desoto!packard!edsel!bentley!hoxna!houxm!mtuxo!mtunh!mtung!mtunf!ariel!vax135!timeinc!phri!pesnta!hplabs!ames!aurora!al
From: al@aurora.UUCP (Al Globus)
Newsgroups: net.space
Subject: Re: Shuttle Computers and STS power consumption
Message-ID: <372@aurora.UUCP>
Date: Wed, 26-Jun-85 23:03:39 EDT
Article-I.D.: aurora.372
Posted: Wed Jun 26 23:03:39 1985
Date-Received: Tue, 2-Jul-85 06:28:29 EDT
References: <2322@mordor.UUCP>
Organization: NASA Ames Research Center, Mtn. View, CA
Lines: 24

> 
>         A recent posting about  shuttle  cooling  reminded  me  of  an
> article  I  ran  across  in  an  old (1980) IBM document.  The article
> described the onboard computers as System/4Pi model AP-101 processors.
> The  AP-101  appears  to  be  a  3  KIP machine with an intensely ugly
> architecture.  Main memory was described as 104KW  (36  bit)  of  core
> (really  core!)  with  400  ns access time.  What startled me were the
> physical characteristics.  Each of the five  "processors"  appears  to
> consist  of  two  boxes.  The  boxes  weigh  58.9  pounds and draw 350
> Watts!!!!! This would imply a system total of almost  600  pounds  and
> 3500  Watts.  Did  they actually use this refugee from the Smithsonian
> on the Shuttle?

I'm almost positive the answer is yes, they do and will continue to
do so for the forseeable future.  If it's not that particular
model, it is a close relative.  Sickening isn't it?

Software note: software development for launch is supposedly very close
to the critical path and definitely must see major improvements to support
24 launches a year.

More horrors: there are some 250 microprocessors on the shuttle with a
variety of incompatable languages, operating systems, and development
systems between them.