2 Nov
2016
2 Nov
'16
10:10 a.m.
On Tue, 11/1/16, Jon Elson <elson at pico-systems.com> wrote:
? Also, some IBM publications (where I'm more
familiar with their models) had some photos
of machines that probably were in-house
prototypes that were quite different than the
production version.
Along the same lines, the picture in the original PDP-8
manual was of a machine that had a front panel that
looked more like the PDP-5 panel than the one shipped
on the 8s. Given how close the machines were in
architecture, it wouldn't be surprising for a prototype.
As it turns out, I saw the picture in the manual a few
years before I ever saw a real straight-8. To this day,
the real straight-8s look a little "wrong" to me.
BLS
2 Nov
2 Nov
6:20 p.m.
On 11/02/2016 12:10 PM, Brian L. Stuart wrote:
On Tue, 11/1/16, Jon Elson <elson at
pico-systems.com> wrote:
Yup, in fact, I think you could turn a PDP-5 into a PDP-8
with about 25 boards
(one tray of system building blocks) and some wire. Mostly,
add an electronic
instruction counter register and alter the instruction fetch
logic to stop accessing
location zero. That's the only difference visible to the
programmer, I think.
Jon
Also, some IBM publications (where I'm
more
familiar with their models) had some photos
of machines that probably were in-house
prototypes that were quite different than the
production version.
Along the same lines, the picture in the original PDP-8
manual was of a machine that had a front panel that
looked more like the PDP-5 panel than the one shipped
on the 8s. Given how close the machines were in
architecture, it wouldn't be surprising for a prototype.
As it turns out, I saw the picture in the manual a few
years before I ever saw a real straight-8. To this day,
the real straight-8s look a little "wrong" to me.
BLS
8:39 p.m.
On Wed, Nov 2, 2016 at 7:20 PM, Jon Elson <elson at pico-systems.com> wrote:
Yup, in fact, I think you could turn a PDP-5 into a
PDP-8 with about 25
boards
(one tray of system building blocks) and some wire. Mostly, add an
electronic
instruction counter register and alter the instruction fetch logic to stop
accessing
location zero. That's the only difference visible to the programmer, I
think.
And changing the interrupt location, which IIRC was 0001 on the PDP-5.
2971
days inactive
2972
days old
2 comments
3 participants
participants (3)
-
blstuart＠bellsouth.net -
elson＠pico-systems.com -
spacewar＠gmail.com