Re: non sequitur

["Robert I. Eachus" <rieachus@comcast.net>]

Dennis Lee Bieber wrote:

> 	My condolences <G>

Why?  When we were validating the Ada compiler, I spent some time doing 
tuning on a 9x system (DPS6-96 I think).  But aside from that I did all 
my development work on Multics.

I really feel sorry for anyone who had to use a Level-6 or DPS-6 system 
running Mod 400 that one of a (very) few people hadn't set up for doing 
so.  The 'normal' settings of Mod 400 made it dog slow as a development 
system.  I went out to Minneapolis to help do that for the system some 
Ada people used.  If you knew which parameters to twiddle you could get 
impressive performance out of Mod 400.  We ran the entire Ada validation 
suite in around 4 hours on the 9x system, and we ran it on the whole 
line during the official validation, although it took a lot longer on 
the DPS-6/22 and /3x machines.

> 	It worked nicely -- until it got warm. Then the circuit card
> would sag, breaking contact with the edge connectors, and leaving the
> mainframe thinking it had idle connections on one side, and lots of
> terminals thinking they had slow connections on the other...

It sounds like the Level 6 you had was installed outside a climate 
controlled environment.  When Honeywell had some initial problems on a 
system for the Navy, they found out that probably 95% of Level 6 systems 
were installed in the same room as a climate-controlled mainframe. If 
you had one of what we called the 5x series processors, you could have 
exactly that problem.  The DPS-6 system that replaced it was fine, but 
for the 9x systems climate control was made a requirement.  (Of course, 
they were really 32-bit mainframes with the 16-bit Level-6/DPS-6 
instruction set supported, about equivalent to a VAX 11/785.)

A couple of other funny (or nasty now) stories.  When we did the Ada 
compiler validation, we were the first people _ever_ to run a DPS6 or 
Level 6 machine under Mod 400 24/7 for more than a week.  We know this 
because we found several memory leaks in Mod 400 and patched them.  One 
particularly nasty leak would lose 2 bytes of physical memory each time 
a new process was started.  I particularly remember that one, because I 
had to turn the patch in on punched cards! (Three of them.)  The last 
time I ever used a card punch, and the only time during my 5 years at 
Honeywell.

One other "high priority" project for the Ada group was to get the 
process for submitting software to Testing and Configuration moved into 
the modern era.  The first program submitted that was written in Ada 
came back with two complaints: No copyright string in the executable, 
and no area reserved for patches!  We added two pragmas to the compiler, 
but by the time I left, we had killed the patch area requirement for 
products written in high-level languages, and OS patches could be 
submitted on floppy disk, among other things.

-- 

                                           Robert I. Eachus

The ideology he opposed throughout his political life insisted that 
history was moved by impersonal tides and unalterable fates. Ronald 
Reagan believed instead in the courage and triumph of free men and we 
believe it all the more because we saw that courage in him.  -- George 
W. Bush June 11, 2004