[Info-vax] OT: obscure PDP11 OSes (even more dinosaury)

[Johnny Billquist]

On 2015-06-16 21:29, glen herrmannsfeldt wrote:
> Dennis Boone <drb at ihatespam.msu.edu> wrote:
>>> With the 11/780 being the first VAX running VMS I used, I would agree.
>>> It was a dog.  And we won't even go into what happened to performance
>>> when some Lieutenant decided to compile something in Ada.  Was a good
>>> time to go to lunch, or in most cases just go home for the day.
>
>> The 780 wasn't _that_ bad.  I worked on one that tended to have 60+
>> interactive users, and two or three batch compiles, all going at the
>> same time.  It was certainly slow at that point, but that's a hell
>> of a workload for a 1MIP machine.
>
> Many of them might have had less real memory than appropriate
> for the use and load.  With enough real memory, the processor is
> plenty fast enough.

The 11/78x was not bad, but it certainly had some performance issues. 
The main one being that memory was sitting on the same SBI bus as 
everything else, making that bus saturated.
One of the biggest improvements of the 8600 (anyone remember 
VAX-11/790?) (which was the first follow on to try and improve the 
performance from the 11/78x) came from just moving the memory away from 
the SBI, and putting it on a separate bus.

	Johnny