🚀 go-pugleaf

In article <17496@glacier.STANFORD.EDU>, jbn@glacier.STANFORD.EDU (John B. Nagle) writes:
> 
>       I'd like to encourage people to write simulators for one or two of the
> early machines, as a way of keeping the history alive.  The ENIAC was 
> plugboard-programmed, so it is not an enormously interesting machine to
> simulate.  But simulating IAS, or Binac, or Whirlwind, would be a useful
> exercise.  Especially if some original software could be found and brought
> back to life.

Actually the ENIAC is quite fascinating to simulate.  It has the most important
feature a computer can have -- a neon bulb for each flip flop (connection 
machine -- eat your heart out).  Also, it was only plugboard-programmed until
1948 at which time it became the first stored-program computer (although
the store was read-only).  Prior to 1948, it was a parallel computer
so there are doubtless many who wouldn't want to simulate it due to the 
difficulty of matching it's speed.  

The biggest problem in simulating it is getting suitable documentation.  
All the ``good stuff'' never made it to the journals, but lies buried
in various technical reports (mostly from the Moore School at U Penn
and BRL at Aberdeen).  Of course, with a little imagination one can
get fairly close.

The earliest electronic stored-program computers that are
well-documented in the public literature seem to be the EDVAC
(proposal in Von Neumann's collected papers as well as significant
discussion in the Moore School Lectures reprinted by MIT Press) and
the ACE (Turing's proposal reprinted by MIT Press -- which differs
from the machines actually built under that name). 

A number of ISP descriptions of slightly more recent machines are available
in Siewiorek, Bell, and Newell's (1982 successor to 1971 Bell and Newell)
Computer Structures - Principles and Examples.  I have also heard that some
other author did a book of ISP descriptions, but I haven't seen that book.

----- BOB (webber@athos.rutgers.edu ; rutgers!athos.rutgers.edu!webber)