1.1 Various small items were changed. (The unlikely sequence ESC ESC Z% Z% used
to crash the system). “Delete line and scroll up” now cleared the bottom
line of the display if the cursor was there. (Previously nothing
happened.) Cursor addressing, Set, Reset, and Test now accepted nested
escape sequences during the coordinate string.
1.2 A few small internal changes made. The input buffer (used to hold
incoming characters while the screen is being scrolled) was reduced to 64
bytes from 128 bytes, (128 bytes was a bit of a waste), and the internal
stack area was increased.
2.0 Support for the Rotec keyboard and programmable function keys added,
hence the jump to 2.0. The monitor size now exceeded 2k and required a
2732 rather than the 2716 of the earlier releases.
2.1 The current release of IVC-MON. Sections of the monitor code were
restructured to give a faster response to keyboard poll requests. This
has a noticable effect on the screen update rate for operating systems
such as CP/M where the BDOS polls the keyboard everytime it outputs a
character. Another Escape sequence was added to allow lines to be
“locked” at the bottom of the screen.
(3.07) Although the IVC has now been discontinued one further update of the
software is envisaged. (When time permits – don’t expect it tomorow!)
This will include SOME of the features that have appeared in the SVC
monitor (see below).
With a new design on the drawing board Gemini was faced with the usual
decision – Do they use a “state-of-the-art” device and design a system around
that and accept the attendant constraints? (e.g. the in-built character set
and screen formats offered by some graphics controllers.) Or should they
remain as far as possible compatible with the IVC? They – wisely in my view –
opted for the latter case.
So following the IVC came the SVC. Why SVC? well they had to call it something
and it does bear some relationship to the IVC. In fact it is upwards
“compatible with the IVC. (i.e. You can replace an IVC by an SVC and 99.9% of
programs will continue to run as before.) So what do you get by moving from an
"I" to an "S"? Well the Z80A changes to a Z80B (6MHz operation), the screen
memory goes up in size from 2k to 8k (although some of it doubles as the
character generator), a 256x256 pixel graphics mode appears, a 40 x 25 display
option is provided (the 0.01% incompatibility), a UART can be added on a
piggy-back board to make the SVC a stand-alone terminal, a range of attributes
appears, a buzzer is added, a Gemini serial keyboard can be used and finally
you get ESP.
I think that is about enough for now, and so I’m afraid you will have to wait
until the next issue for part II to find out about the ESP.
1. PARKINSON D.W., "Remote Terminals and the Galaxy", 80-BUS NEWS 1-4 Nov-Dec