Re: Why VAX Was the Ultimate CISC and Not RISC

On 3/2/25 5:27 PM, John Levine wrote:

According to BGB  <cr88192@gmail.com>:

I had thought it apparently used a model similar to the 65C816.
>
Namely, that you could address 64K code + 64K data at a time, but then
load a value into a special register to access different RAM banks.

 Not really.  The low end PDP-11's were 16 bit, 64K was it.
 The larger ones had memory mapping with 8K pages, a size carefully
chosen to be too large for paging, but too small to map whole programs.
There were three modes, user, supervisor, and kernel, with 64K instruction
and data in each.  The kernel changed the maps by poking values into
I/O addresses, so it's not something a normal program could do.
 Unix only used user and kernel so for our early bitmap terminals, I
mapped the screen's video memory into supervisor data and set the
mode bits so you could access it with MOVE TO/FROM PREVIOUS DATA
SPACE. C didn't generate those so we had some little assembler
routines.
 Given the way the PDP-11 was set up, it's hard to think of a memory
expansion scheme that wasn't a grotesque kludge so I think it was
the right decision for VAX to have a new instruction set with a mode
to run PDP-11 code, sort of like the 386's virtual 86 mode for
real mode 8086 code.
 

That was not what customers were interested in.  There were various
Unix variants available for the PC, but the customers preferred using
DOS, which was preinstalled and did not cost extra. ...

 Yup.  PC/IX was a really nice Unix port for the IBM PC and nobody was interested.
 

As this (the kernel part) was my project, it was very disappointing.  I think IBM priced such that with DOS being "free", it had no chance.
Brian