Re: Does an enhanced //e _ALWAYS_ have 80 col. support?

["Craig Bower \(Bender MX\)" <NOSPAMnitroacd@telusplanet.NAYSPAM.net>]

"Oliver Schmidt" <ol.sc@web.de> wrote in message 
news:co35tg$bl5$1@online.de...
> Hi Craig,
> Hopefully this makes my two questions clearer:
> 1. Is it okay to read the 80 col softswitch ($C01F) for detecting 80
> col mode even if there's no 80 col card installed at all?

Hi Oliver,

It can't hurt, but won't the softswitch just show you the same thing if
a) the card isn't installed and b) the card is installed and not active?

> 2. How do I detect that an 80 col card is actually installed? Checking
> for the existence of the $C300 firmware seems to be wrong from what I
> understood from your answers.

Not specifically wrong, I misunderstood your original question! :)

Look for specific "signature" bytes that are in the $C300 range and are
unique to the 80-column card.  From my somewhat limited experimentation
tonight, the old style (BIG) and new style (small) 80-Col / 64K cards
both have the first three bytes as follows:

C300- 2C 43 CE

And when I check the same memory location after pulling the cards,
I get the exact same bytes.

Your best bet?  Grab a real //e power it off, pull the 80-Col Card
and dump C300.C3FF to the printer.  Next install the 80-Col Card
and dump the C300.C3FF range again.  Now if you have a bunch
of different year/version 80-Col Cards, try them all.  Compare the
differences between the ROM dumps and see if there's anything
significantly different between the filled slot 3 and the empty slot
3 hex dumps.

If that proves futile, I'd imagine you could always write a
routine that *attempts* to write into bank switched memory on the
80-Col Card and then double checks main RAM to ensure it's not
there?  You'd need to write to an area of the 80-Col Card that
isn't part of the 64k extended, in case the user was running just
the 80-Col Card and not the extended version.

Best of luck,
Craig