I tried to pull those monitor signals high, however the effect is the
same. Guess I have to look closer at the memory defects. One way to go
is to use a dedicated test rom or one of the exerciser cartridges. The
test rom was used by the service personnel only, and therefore will be
hard to find. However I recently acquired a system 45B/C exercise
cartridge. The tape contains a couple of memory checks among other tests
and supports autostart.
Since I assume you've got a running system, can you tell me whether the
system tries to launch the autost program from a 9845 tape before,
during or after the power-on memory test (in case the "auto st" key is
latched, a tape is inserted and the system is being switched on)? If the
program is read from tape _after_ the power-on test has been completed,
I'll have to find another way.
--Ansgar
Yes.
In the 98780, there is one video timing chain, physically located on the
text (alpha) PCB. It's the traditional chain of counters, decoded by AND
gates, used to control JK's to provide various timingt signals (you need
the schematics, OK :-))
Many of the signals are passed across the monitor backplane to the
procesor (graphics) PCB, but there is no complete timing chain on that
PCB. The only board that provides syncs to the analogue section is the
text PCB, and from what I rmember, tbere are no control inputs to that
timing chain depending on whether it's in text or graphics mode.
I was under the impression that the 98770 (colour) and 98780 (enhanced
mono) were pretty similar to the user apart from the obvious difference.
-tony