Re: C64 MMU POC

From: Mia Magnusson <mia_at_plea.se>
Date: Wed, 9 Oct 2019 13:54:02 +0200
Message-ID: <20191009135402.00005471_at_plea.se>
Den Sun, 22 Sep 2019 14:57:19 -0500 skrev Jim Brain <brain_at_jbrain.com>:
> On 9/22/2019 2:37 PM, Mia Magnusson wrote:
> >
> > Perhaps use extra addresses in the VIC-II address space? (Not sure
> > if anything bad happens if you touch those addresses though).
> Can you write to RAM under VIC-20 registers?  Remember, the unit sits
> in the RAM sockets, so it only gets writes if RAM is written.

VIC-20? :O

I assume you refer to VIC-II (and in general the 4k I/O block on a
C64). I haven't done any testing myself but I would be really surprised
if writes bleed through. The rationale for writes to rom to bleed
through to ram must be that the writes need to go somewhere and it's
a handy feature. (Btw compare with the copy protection schemes that
when detecting a pirate copy apparently damaged ROMs in some 1541 drives
by writing to ROM until the ROM chip broke - I assume this would be
impossible in a C64 thanks to that the writes get redirected to RAM).

> > Maybe you could put registers in the $FFxx range like on the C128,
> > with enable/disable in the regular I/O space.
> 
> I think that would work, as writes bleed through to RAM with ROMs
> banked in, but reads would not work, so it'd be best to use this for
> the "quick, set my mapping, no need to read" use case.  I'm not sure
> folks want to bank out ROMs all the time to read the MMU config, but
> could be wrong.

In theory you wouldn't need to be able to read any MMU registers at
all, as software could keep copies of whatever it needs or rewrite them
to whatever state is needed.


-- 
(\_/) Copy the bunny to your mails to help
(O.o) him achieve world domination.
(> <) Come join the dark side.
/_|_\ We have cookies.
Received on 2020-05-29 21:06:48

Archive generated by hypermail 2.3.0.