Re: Strange ASCII character mixup in my iigs

[dempson@actrix.gen.nz (David Empson)]

Wolfgang Kemper <wkemper@work.com> wrote:

> I tried on the weekend the Shake and Roll and also watched for "dead 
> Bugs" bud it has not helped. I lifted all IC's with a socket to clean
> the pins but also no success.
> 
> I also have to correct myself, I tried to play around with the ASCII 
> code and figured out that even if I see the "#" the "a" is recognized.
> (So wrong statement that "c#t" will not be recognized, sorry, do not 
> know what I have done wrong wen I tried it)

Your IIgs appears to be suffering from a corrupted address getting into
its character generator ROM, though it is a rather weird one: an $E1
charcter ("a") is being mapped to $A3 ("#").

Does it happen with inverse text (in 80-column mode) or just normal
text? Try this (and yes, that's a real hash symbol in the first line):

PR#3
INVERSE
PRINT "Bad"

Do you get an inverse "Bad" or "B#d"?

Does it happen if you select another display language in the Control
Panel?

I wonder if lo-res graphics mode is affected similarly? That's a bit
harder to test.

In the IIgs, the characters displayed in text mode are supplied by a ROM
which is located inside one of the ASICs: probably the Mega II (it could
be the VGC). It isn't likely that a hardware fault on the motherboard
could cause a problem like this - it is more likely to be a fault within
the chip itself, and is therefore unfixable short of replacing the chip
(which isn't practical for a surface-mount chip).

> Anyhow under Geos it will not happened, a lower case a is a lower case a.
> Is GEOS using its own character routines?

Did you mean "Geos" or "GS/OS"?

GS/OS mostly runs in graphical mode and uses software fonts to display
text, so it wouldn't be affected by this problem.

I'm only vaguely aware of Geos. Assuming it has a graphical user
interface, then it also would be using software fonts.

The built-in character set is only used when the screen is in text mode.

-- 
David Empson
dempson@actrix.gen.nz