Re: Historical evolution of CPU perf

On 2024-10-12, Michael S <already5chosen@yahoo.com> wrote:

On Fri, 11 Oct 2024 14:40:59 GMT
scott@slp53.sl.home (Scott Lurndal) wrote:
>

Sarr Blumson <sarr@sdf.org> writes:

MitchAlsup1 <mitchalsup@aol.com> wrote: 

On Wed, 9 Oct 2024 19:18:40 +0000, Kerr-Mudd, John wrote:
 
I remember the PDP-11/20 in the computer lab at NCR.
Last person out at night would flick the power switch OFF, and
the computer was OFF in 1/60 of a second.
First person in would flick the switch ON and the computer was
back where it was turned off in 1/60 of a second.
 
We used the 11/20 as a remote debug device for the 8085 cash
register machine(s) we were building. 

>
Core memory: slow to access but also slow to forget. 

 
Compared to the alternatives at the time, it was the fastest
gun in the west.
 

>
At the time of 11/20 debute (1970) it already was not.
Custom SRAM was in use for 3-4 years and first OTS SRAM (Intel  3031)
was shipping for something like a year.

In my embedded/datacomm world, I did not meet solid state RAM until fall
1980. All the PDP-11/20, PDP-11/05, PDP-11/10, PDP-11/35, and
PDP-11/45's were still core at that time, and the first system with
solid state memory that I worked on had a bad bug in the memory, which
soured me on solid state for a couple of years more.

Essentially, allocation bitmaps (especially disk allocation bitmaps)
where you set a bit after it had been 0 for a long time were prone to
revert after it had not been looked at for another hour. I figure that
it must have been SRAM.