Re: WDC's 32-bit W65T32

[Mike Willegal <mike@willegal.net>]

The bus itself isn't the issue, it's the addressing modes of the 
processor which must generate more bits for the addressing the wider 
bus.  Intel accomplished this by adding segment registers, which are 
used to add to base the address of the instruction and data fetches. 
Special instructions must be added to manipulate the segment registers. 
 Once the segment registers are set up, you can run old code that is 
not aware of the segment registers anywhere in your expanded address 
space.

Regards,
Mike Willegal



aiiadict@gmail.com wrote:
> On Nov 2, 8:57 am, Mike Maginnis <magin...@computist-project.net>
> wrote:
>
> > Apparently, this is old news, but it looks like WDC's forthcoming
> > 32-bit  chip is completely 6502/65816 compatible.  I don't know much
> > about CPU architecture, so I don't know if this would matter to the A2
> > community.  Would such a chip impact us at all with, say, the
> > development of new peripherals, etc?
>
>
> It's been rumored to be in development for over ten years, I think...
>
> compatible with 6502...  ugh...  how do you get 32 bit's address with
> only 16 address lines?  Some sort of scheme to put first 16bits on, ?
> latch it?
> in another chip, then get next 16 bits...  then an MMU would address
> the appropriate RAM chip?
>
> If it doesn't happen, how about modifying a 65816 VHDL or verilog
> core?
>
> would be cool to get it running, and code a C compiler that works with
> 32bits... we could port a ton of software to the II...  except for the
> low
> resolution of graphics and speed of CPU...
>
> Rich