(j3.2006) (SC22WG5.4208) RE: RE: 43 Fortran compilers

[dick.hendrickson at att.net]

  -------------- Original message from Robert Corbett <Robert.Corbett at Sun.COM>: --------------


> Walt Brainerd wrote:
> > Just to add to the confusion:
> >  
> > [the CDC 1604] Cray's first design, perhaps, and
> 
> Perhaps not.  The ERA 1103 is usually listed as Cray's first
> design.

Ahh, the good old days!!

I think Robert is correct, Semour built the CDC machines after he
left ERA, which (I think) went on to become UNIVAC.   

The CDC "factory" where the 1604 was built was located at 501 Park 
Avenue in Minneapolis.  An interesting numerologic fact.  

For what little it's worth, I was actually in that building and used a CDC
160  computer to do real physics.  The 160 was a true desktop computer
(being roughly the size of a desk) and had 4096 12 bit words.  The Fortran 60
compiler was a three pass compiler.  Each pass of the compiler was 
loaded in from punched paper tape and the intermediate pass-to-pass 
data was also on punch paper tape.

Later on, I (and another physics grad student and a senior computer center 
person)  heavily modified the original CDC 1604 Fortran compiler for the 1604
to produce the U. of Minnesota Fortran compiler.   A fast compiler with
excellent (for the times) diagnostics that generated near perfect code if
you didn't use DO loops.  Being a line at a time compiler it was a little
weak on constructs that extended over more than one line.   

Dick Hendrickson


> 
> Bob Corbett
> _______________________________________________
> J3 mailing list
> J3 at j3-fortran.org
> http://j3-fortran.org/mailman/listinfo/j3
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://j3-fortran.org/pipermail/j3/attachments/20100302/d45b96e0/attachment.htm>