>From owner-simulators  Fri Sep 29 13:45:52 1995
Received: from meitner.cs.washington.edu (meitner.cs.washington.edu [128.95.2.104]) by june.cs.washington.edu (8.6.12/7.2ju) with ESMTP id NAA07215; Fri, 29 Sep 1995 13:45:51 -0700
Received: from localhost (localhost [127.0.0.1]) by meitner.cs.washington.edu (8.6.12/7.2ws+) with SMTP id NAA24140; Fri, 29 Sep 1995 13:45:48 -0700
Message-Id: <199509292045.NAA24140@meitner.cs.washington.edu>
To: rtcg@cs, simulators@cs
Subject: [von Neumann 45], [Stern 81] First Draft of a Report on the EDVAC
Date: Fri, 29 Sep 1995 13:45:47 PDT
From: "		    pardo@cs.washington.edu" <pardo@meitner.cs.washington.edu>
X-Message-Id: simulators@cs.washington.edu, message #1995-09-001
X-Unsubscribe: e-mail `majordomo@cs.washington.edu', body `unsubscribe simulators'
X-URL: `http://www.cs.washington.edu/homes/pardo/sim.d/mail.d/index.html'

%A Nancy Stern
%T From ENIAC to UNIVAC \- An Appraisal of the Eckert-Mauchly
Computers
%I Digital Press
%D 1981

%A John von Neumann
%T First Draft of a Report on the EDVAC
%I University of Pennsylvania
%D 30 June 1945
%O Appears in [Stern:81], pp. 177-246
%X Build a computer starting from fairly basic principles.
  (pp. 218-219) ``We shall fix the size of such a number at 30 binary
digits [[bits -Pardo]] ... To this must be added one unit for its sign
... and it is advisable to add a further unit in lieu of a symbol
which characterizes it as a number (to distinguish it from an order
...''  I.e., a code/data tag bit.  It's not clear to Pardo that the
distinguishing bit is used.
  (pg. 237) ... discusion of control transfer instructions ... ``[A]
transfer might provide [that control] should return to its connection
to the [memory location] that fillows upon the one containing the
transfer order [or that control] should continue with that [transfer
location] and accept order from there on in thatural temporal
sequence.  It is conventient to call a transfer of the first type a
_transient_ one and one of the second type a _permanent_ one.  It is
clear that permanent transfers are frewquently needed ... it seems
very doubtful whether [transient transfers] are ever needed in true
orders [and] can always be expressed by two permanent transfer
orders.''   In short, the EXECUTE instruction!
  (pg. 240) Description of instruction encodings; notes the
``code/data'' bit.  (pg. 242) ``immediate'' operands.  (pg 244) Notes
that ``code/data'' bit again.
  (pg. 243) Instruction words are 31 usable bits  but only about 13
are used (instructions with immediate operands occupy two words).  It
might be possible to e.g., pack two instruction pser word but doing so
should be avoided if it ... would slow overall execution ... and would
save only the part of memory used for instructions, which is typically
small compared to the data portion.  Packing should only be done if it
simplifies the logical structure of the code.  [[RISC predates the
first working programmable-store computer! -Pardo]]