Felix,

I don't assume anything especially since this thread can be ref'd ad hoc in the future in a differenct context.

Yes the decode i indicated "ghosts" the 8255's this could be fixed using the alternate '138 if you felt it mattered.

I couldn't tell if you had an argument with the 32k window into the SRAM. It's just a conveniant way to address memory (i use 512 SRAMs this way) with I/O in place concurrently.

regards,
p