http://archive.6502.org/datasheets/wdc_ ... r_2000.pdf

I got bored and decided to read an old 65816 datasheet- I'm curious if the '816's operation changed in any manner between 1981 and now besides increases in clock speed. For the most part, the old datasheet seems very similar to the current 2010 one. One thing I did notice, however, that is in the 2000 datasheet and not the 2010 one is the following statement on page 54:



One thing that's (rightfully so) brought up in most '816 threads on this site is that the control logic is slightly more complicated than the 6502, and one should use the VPA/VDA outputs to figure out what's going on before accessing memory or I/O. As someone who doesn't have an Apple II, does anyone know and/or remember to what the above quote refers? It seems like one would run into trouble if I/O accesses weren't qualified .

Since Apple II was discontinued 10 years before that datasheet, I wonder if it's simply a remnant from even older datasheets that was never removed.

Sounds like some Apple disk hardware made use of the extra accesses which sometimes occur. That's not too surprising, if Woz was involved. But it does mean extreme compatibility is called for.

Correct. If VDA and VPA are ignored then the bus gyrations of the 65C816 during intermediate cycles of an instruction are like those of the NMOS 6502. The Apple disk hardware was dependent on that to function.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Specifically, some of the "dead" cycles exist in compatibility mode specifically for the Disk ][ card and similar. IIRC it was something like the extra cycle on an RMW instruction or a page-crossing penalty or similar. The Disk ][ card is amazing, but some of the software games played are a bit much.