> > How ... or does ... Aaron's continued efforts to port parts of the code to C++ fit
> > into supporting multiple machines?
> I may be way off base, but I see it as mostly "updating the architecture to make use
> of modern compiler features" - in other words, just to update. That it may benefit
> running multiple machines simultaneously is "just a nice side effect."
Not to make light of the C programming community, but I think at this point, there's more active C++ programmers than C programmers. By moving to C++, I think the MAME source becomes more accessible to more of the programmer population and hopefully attracts more people to submit source contributions and fixes.
Just because someone is a good C++ programmer and C++ is a super-set of C, doesn't mean that that person is necessarily a good C programmer.
C is kind of a lowest common denominator thing, so the down side of moving to C++ is potentially giving up the ability to easily port MAME to older architectures that have poor (or no) C++ compiler support. However, in practice I don't know how much that really occurs. Most hardware that doesn't have a C++ compiler isn't going to be able to run the modern MAME codebase anyway. You'd be better off porting from an old C version of MAME and back-porting ROM updates and fixes on a case by case basis.
GroovyMAME support forum on BYOAC