Hallo Andre, > This seems reasonable. Open inputs normally read as high anyway (due to > internal pullups) Then explain me the $DE00/$DFFF trick of Marko :-) In this case the databus reflects something originated by the VIC. But there is an old saying here in the Netherlands: "Baadt het niet, schaadt het niet" (If it does not help, it does not hurt either). > Hae? changing the mapping in the middle of the write will cause a > 'half-write' (or what you like to call it) to the newly mapped address? > I don't think I get the point. The only thing I do is disabling the mapping. This means that the output does NOT reflect one of the registers. So I safely can change them. > Hm. You should probably think of mapping the addresses $000000-$0fffff > (1MByte) > with the MMU, i.e. with slow access. Then you can have some additional > RAM in this space to replace the ROMs for example, which you couldn't do > otherwise (without sacrificing 8032 RAM for this purpose). I think you missed the point. A16-A23 of the CPU and those of the MMU are placed PARALLEL! Which lines are active depends if the 65816 is in the area $000000/$00FFFF or not. This means I can replace ROM with RAM form every other big page I want. > See out Phi2 discussion. I would rather use bus driver to decouple the > busses. You should at least mention that, I think. Thanks, forgot to correct this one. Groetjes, Ruud
Archive generated by hypermail 2.1.1.