"Dynamic Address Translation" is an impressive phase, but early
> machines had nothing more than a relocation register and a protection
> register. The user's memory addresses were summed with the relocation
> register to get a physical address. The protection register set an
> upper bounds on the user's memory access and caused a trap or fault to
> the kernel if exceeded.
Funny you should mention that, IIRC a KA-10 processor (first of the DEC10 line) had two protect/relocate register sets. One for the 'high' segment, usually pure code, and one for the 'low', mixed code and data. I did spend some time playing with KA-10 s/n 9, which had only one set. In the early days, the registes were optional, as was the byte-pointer and floating point hardware.
I disagree. I installed a fully functional 8-user timeshare system
> on a PDP 8 with 12k words of memory back in 1972. The early
> Unix and Decsystem 10's were amazingly efficient for the resources > available then.
Yes. A DECsystem10 KA-10 processor would happily do 40+ users with
160kwords of -core-. That's using a 1Mword drum as swap device and with about 44k resident as the monitor (o/s). Of course, it wasn't a speed demon compared to systems ten years later, but for the time, it was great.
Also, since a "batch process" effectively means performing some processing based on pre-written instructions, as opposed to interactive instructions, a PDP-8 could certainly do them. Not fast, mind you, but it could.
z!